P O L S K A
KOMITET
A K A D E M I A
NAUK
N A U K
O R I E N TA L I S T Y C Z N Y C H
ROCZNIK
ORIENTALISTYCZNY
ukazuje się od 1914/1915 r.
TOM LXXI
ZESZYT 2
WARSZAWA 2018
Wydawca:
Komitet Nauk Orientalistycznych Polskiej Akademii Nauk
Komitet Redakcyjny:
Marek M. Dziekan (redaktor naczelny), Marta Woźniak-Bobińska (sekretarz redakcji),
Jaakko Hämeen Antilla, Agata Bareja-Starzyńska, Eduard Gombár, Lidia Kasarełło,
Agnieszka Kozyra, Ewa Siwierska, Lidia Sudyka, Gábor Takacs
Rada Redakcyjna:
Janusz Danecki (Warszawa), Edward Lipiński (Bruksela), Alfred F. Majewicz (Toruń),
Piotr Taracha (Warszawa), Przemysław W. Turek (Kraków),
Vladimir Uspensky (St. Petersburg), Witold Witakowski (Uppsala)
Adres:
“Rocznik Orientalistyczny”
Faculty of Oriental Studies
University of Warsaw
00-927 Warsaw 64
e-mail: rorient@pan.pl
© Copyright by Komitet Nauk Orientalistycznych PAN and Dom Wydawniczy Elipsa
ISSN 0080-3545
ISBN 978-83-8017-238-8
Nakład 200 egz.
Realizacja wydawnicza:
Dom Wydawniczy ELIPSA
ul. Inflancka 15/198, 00-189 Warszawa
tel. 22 635 03 01, 22 635 17 85
e-mail: elipsa@elipsa.pl, www.elipsa.pl
Spis treści / List of Contents
Preface
Ákos Bertalan Apatóczky, Early Mandarin Profanity and Its Middle Mongolian Reflection in
the Vocabulary of the Wu Bei Zhi 武備志. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Viacheslav Chernev, On the Use of Past Participle Forms in Oẓon-oẓaǩ bala saǩ (“Quite
a Long Childhood”) by Mostay Kärim . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Gohar Hakobian, Landscape Terminology in Western Iranian . . . . . . . . . . . . . . . . . . . . . . . . . .
Murat Işik, Oghuzic and Kipchak Characteristics in the Book of Leviticus, Gözleve Bible
(1841) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Stanisław Jan Kania, Śāntarakṣita and Kamalaśīla on the Cārvāka/Lokāyata aphorism: ‘from
these, consciousness’ (tebhyaś caitanyam). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Shahla Kazimova, The Lost Legacy.The little known heritage of Azerbaijani emigration literature
in Poland on the example of Mehemmed Emin Resulzade’s works . . . . . . . . . . . . . . . . . . . .
Patrycja Kozieł, Oral Literature and Indigenous Knowledge: The Case of the San People from
Southern Africa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Petra Maurer, Lexicography of the Tibetan Language with Special Reference to the “Wörterbuch
der tibetischen Schriftsprache” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Michał Németh, A Historical Phonology of Western Karaim. The Process of Its Diversification
into Dialects. Part 2. Supplementary Data on the Absolute and Relative Chronology of
Sound Changes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Jamila Oueslati, The Valued Beauty of Ḡnā. A Genre of Tunisian Women’s Songs . . . . . . . . . .
Lidia Sudyka, Kerala Women’s Writing in Sanskrit: Ambādēvi Tampurāṭṭi of Cemprol Koṭṭāraṁ
– Her Life and Literary Oeuvre. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Mark Turin and Benjamin Chung, Colour Terms in Tibeto-Burman Languages. . . . . . . . . . . . .
Vladimir Uspensky, About an Attempt to Use the Cyrillic Alphabet for the Mongolian
Language . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Anna Zalewska, Forgotten Jewels: Japan in Poetry and Prose Written by Polish Authors
until 1939 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7
9
39
57
66
77
94
105
118
146
162
187
198
249
259
Lista Autorów / List of Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
269
Lista Recenzentów RO LXXI, z. 1, 2 / List of Reviewers RO LXXI, nos 1, 2 . . . . . . . . . . . .
270
R O C Z N I K
O R I E N T A L I S T Y C Z N Y, T. LXXI, Z. 2, 2018, (s. 198–248)
DOI 10.24425/ro.2019.127213
MARK TURIN and BENJAMIN CHUNG
Colour Terms in Tibeto-Burman Languages
Abstract
In their handling of colour, Tibeto-Burman languages of the Himalayan region show
multiple lexical similarities to one another as well as apparent influences from more
dominant languages such as Hindi, Nepali, Tibetan, and Chinese. As an understudied
family, Tibeto-Burman languages also serve as an important site to explore modern colour
theory and conceptualisation. Outlier languages in the Tibeto-Burman family that do
not appear to follow either traditional or revised versions of Brent Berlin & Paul Kay’s
theories are of particular significance. This survey provides a systematic review of the
existing literature and a baseline of comparative colour terminology for these generally
vulnerable and often endangered languages.
Keywords: Tibeto-Burman languages, colour terminology, Himalayan studies, loanwords,
reduplication
In this article,1 we describe, review and assess colour terminology in nineteen TibetoBurman languages spoken across the Greater Himalayan range. Through a systematic
analysis of publically available synchronic linguistic data published in various grammars
and articles, this contribution offers an overview of colour terminology in Tibeto-Burman
We are particularly grateful to Sonam Chusang for help in translating and transliterating Tibetan words and
concepts, and to the University of British Columbia and its many libraries, which when taken together, have provided
both funding for this project and made accessible many of the resources that we have consulted. This article was
written on the traditional, ancestral and unceded territory of the hən̓q̓əmin̓əm-speaking xʷməθkʷəy̓əm (Musqueam)
people. In addition, we are thankful to the editors of the journal for their insights and guidance, and the two anonymous
peer reviewers who gave freely of their knowledge and experience. The recommendations and suggestions that we
received have made this article stronger. Needless to say, all remaining errors, misrepresentations and infelicities
are our own.
1
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
199
languages, with a particular focus on languages spoken in Eastern Nepal. We begin with
a brief account of colour theory and recent developments in this important and fastchanging field in order to lay a foundation for the data that follows. We then briefly
describe the Tibeto-Burman language family and note some distinctive features that recur
in our data. In the substantive section of our paper, we review and compare the available
linguistic data, beginning our analysis with more typical, expected colour terminology
and then extending to more complex and specialized inventories.2 Given that relatively
little research has been conducted on colour terminology in Tibeto-Burman languages, and
even less of the available research has been incorporated into modern theory (including
Berlin & Kay’s colour theory), this study serves as baseline for further investigation.
Moreover, this comparative compilation of publically-available linguistic data raises
important theoretical issues in the fields of cognition, linguistic relativism and genetic
relation in an under-described body of languages diverse in phonology, affixation, and
lexicon. This is particularly relevant for the languages that appear to deviate from more
conventional understandings of colour expression.
A Brief Introduction to Colour Theory
In 1969, Brent Berlin & Paul Kay’s revolutionary study of colour perception and
language forever changed how linguists and anthropologists understand the world’s
languages and the worldviews that they transmit. Berlin and Kay’s argument proposed the
existence of universals in human languages for the description of colours, and the authors
argued that over time, a lexicon may develop to incorporate more colour terminology
in a predictable, linear progression.3 A summary of their initial findings is as follows:
1. All languages contain terms for white and black.
2. If a language contains three terms, then it contains a term for red.
3. If a language contains four terms, then it contains a term for either green or yellow
(but not both).
4. If a language contains five terms, then it contains terms for both green and yellow.
5. If a language contains six terms, then it contains a term for blue.
6. If a language contains seven terms, then it contains a term for brown.
7. If a language contains eight or more terms, then it contains a term for purple, pink,
orange, grey, or some combination of these.4
2 In such a typological and comparative endeavor, it is near impossible to standardize transcriptions across
grammars, authors and different theoretical approaches. On consideration, we have chosen to represent the languages
of the region in the original orthographies in which each researcher initially composed their publications. For the
interested reader, we recommend further consultation with the respective grammar and other related sources.
3 Paul Kay and Chad K. McDaniel, The Linguistic Significance of the Meanings of Basic Color Terms,
“Language” (54) 3, (1978), p. 610.
4 Ibidem, p. 613.
200
MARK TURIN, BENJAMIN CHUNG
Later findings relating to the biological processes involved in colour perception as
well as criticism of the original claims have helped to add nuance to, and further develop,
the original theory, including revisions offered by Kay himself. Drawing on Berlin &
Kay’s foundational work, Kay & McDaniel’s analysis suggests that colour universals
result from the neurological aspect of colour reception, challenging positions held by
linguistic relativists. Kay & McDaniel propose that, “each basic colour category can be
regarded as a fuzzy set where the elements in each set are chosen from the set of all
colour percepts,”5 where, “there is a continuous and gradual decline from unity to zero
in the membership values of successive [colours].”6 In their analysis, Kay & McDaniel
challenge the notion of discrete, “semantic [prime] features”7 in colours, and support
the legitimacy of “colours”8 like blue-green. In their revised theory, black, white, red,
yellow, green, and blue are repositioned as the major categories/foci that then combine
and intersect to create more diverse, later stage colours as well as account for perceptive
coolness and darkness.9 Thus, their identification of fuzzy sets validates the inclusion of
non-traditional colours in schematic analyses.
In an additional study, Kay & Maffi expand the theory further and negotiate its
bearings in relation to the newer Emergence Hypothesis (EH), which claims that “not
all languages necessarily possess a small set of words or word senses each of whose
significatum is a colour concept and whose significata jointly partition the perceptual
colour space.”10 They ultimately conclude that Partition (a predisposition to semantically
divide a domain into lexemes) may not manifest in a uniformly typological manner for
colour terms and that it remains a central distinction, as some languages do not exhibit
partition whatsoever.11 A re-ordering of colour partition principles (i.e. general Partition, as
outlined above; Black & White distinction; Warm & Cool distinction; and Red distinction)
derived from Berlin & Kay’s original and revised theories then becomes necessary. Once
incorporated, this theoretical modification can account for great linguistic diversity and
the existence of non-partitioning lexicons.12
In more recent work, Lindsey & Brown explore the universality of colour names
and terminology through the World Colour Survey (hereafter WCS) which had originally
been developed as a database by Kay and emerged out of their landmark study.13 Lindsey
& Brown’s analysis of colour clusters – making use of a similarity metric in order
to more accurately observe patterning – reveals that cluster patterns do approximately
Ibidem, p. 624.
Ibidem.
7 Ibidem, p. 611.
8 Ibidem.
9 Ibidem, p. 637.
10 Paul Kay and Luisa Maffi, Color appearance and the emergence and evolution of basic color lexicons,
“American Anthropologist” (101) 4, (1999), p. 744.
11 Ibidem, p. 745.
12 Ibidem, pp. 746–749, 755.
13 Delwin T. Lindsey and Angela M. Brown, Universality of Color Names, “Proceedings of the National
Academy of Sciences of the United States of America” (103) 44, (2006), p. 16608.
5
6
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
201
coincide with English colour boundaries across multiple languages.14 On account of
the “time-dependent”15 nature of the survey, however, the authors were not able to
determine with accuracy any evolutionary or developmental projections in terminology.
A study by Thierry, Athanapoulos, Wiggett, Dering, Kuipers & Ungerleider of Greek and
English speakers’ colour perceptions supports the proposal that speakers may actually
unconsciously discriminate colours in their lexical inventory.16 Their use of vMMN,
“an electrophysical index of perceptual deviancy detection,”17 could well be evidence
that, “language may fundamentally shape and affect automatic, low-level, unconscious
perception of the experienced world.”18
Despite this flurry of recent interest in colour terminology from researchers in a range
of disciplines, some scholars continue to question the very basis of universal and categorical
claims. Saunders outright refutes Berlin & Kay’s original theory and suggests that, “colour
is not a natural thing (made of reflectances, retinal pigments, opponent processes), but
exists through noticings and reportings as an ensemble of social relations … [and] to
obtain it needs socio-historical and cultural specificities.”19 Unconvinced following
even a rejoinder from Kay himself, Saunders continues to challenge traditional theory,
arguing that, “both empirical analysis and theory involve inherently philosophical and
historiographic endeavors.”20 Such criticism is not altogether unfounded, as the original
theory has been subjected to considerable alterations and ongoing refinements. In addition,
inconsistencies still exist, such as the colour grey being a “wild card at various points
in the sequence,”21 a point conceded by Berlin & Kay.
In this article, we make generous reference to Berlin & Kay’s theory. We believe
that prior acknowledgment of the theory, its critiques and its amendments is crucial for
the analysis of non-Western languages that remain largely under documented. Strangely,
although the World Colour Survey classified 110 languages, not a single Tibeto-Burman
language is represented in the dataset. This presents fertile space for further inquiry,
and challenges – if only through absence – the global reach of Berlin and Kay’s theory
and its putative claims to universality.22 It is widely agreed that in some understudied
language families, Berlin & Kay’s colour patterns appear to surface with regularity.
Ibidem, p. 16611.
Ibidem, p. 16612.
16 Guillaume Thierry et al., Unconscious Effects of Language-Specific Terminology on Preattentive Color
Perception, “Proceedings of the National Academy of Sciences of the United States of America” (106) 11, (2009),
p. 4568.
17 Ibidem, pp. 4567–4568.
18 Ibidem.
19 Barbara Saunders, Revisiting Basic Colour Terms, “Journal of the Royal Anthropological Institute” (6) 1,(2000),
p. 93.
20 P. Kay, In Defense of Color Categories in Thought and Language (Hardin and Maffi, eds.): A Response to
BAC Saunders’s Review, “American Anthropologist” (102) 2, (2000), pp. 321–323.
21 Kay and McDaniel, Linguistic Significance of the Meanings of Basic Color Terms, p. 640.
22 WCS Languages, ISO 639-3 Codes, Families, and Countries Where Recorded, Viewed 16 July 2017, <www1.
icsi.berkeley.edu>.
14
15
202
MARK TURIN, BENJAMIN CHUNG
Despite some major outlier that contain only terms for black, white, and green, in their
colour survey of the Pama-Nyungan language family spoken in Australia, Haynie &
Bowern note that Berlin & Kay’s overall theory does hold.23 Principally, and given that
lexicons do change, Haynie & Bowern investigate how ancestral node reconstruction allows
subgrouping when applied through parameters of potential loss and gain of terms. It is
noteworthy that the loss of colour terms is not theorized by Berlin & Kay as a typical,
observable phenomenon.24 Through a process of reconstruction, Haynie & Bowern offer
a typologically-informed predication of which colours may have historically been lost and
gained along direct linguistic nodes or subdivisions in the wider family. Haynie & Bowern’s
work illustrates how, through processes of both loss and gain, contemporary descendent
languages may have diverged from Proto-Pama-Nyungan, which likely contained only
black, white, and red.25
Discrepancies and differences in the field still exist. Researchers like Wierzbicka
continue to assert that the concept of “colour universals”26 is in reality “self-contradictory,”27
a linguistic oxymoron. Since some languages do not even have a word for colour itself,
no human universal for it can possibly be said to exist. Wierzbicka draws our attention
to what we may call the ethnocentric linguistic bias in the Western study of colour:
languages without some of the colours commonly found in Western languages (particularly
English) are often presented as exhibiting “lexical gaps.”28 It is hard to disagree with
Wierzbicka when she reasons that, “it is not a matter of lexical gaps; it is a matter of
different ways of looking at the world.”29 Wierzbicka asserts that a “natural semantic
metalanguage (NSM)” is necessary to “bridge between the conceptual world of the
linguist and anthropologist and that of the indigenous consultant” and advances a more
culturally-literate, locally-informed and Indigenous perspective to understand colour in
non-Western context.30 As will become apparent below, emic understandings and cultural
groundedness are vital to make sense of colour terminology in the complex and diverse
Tibeto-Burman family of languages.
23 Hannah J. Haynie and Claire Bowern, Phylogenetic approach to the evolution of color term systems,
“Proceedings of the National Academy of Sciences of the United States of America” 113 (48), (2016), pp. 13668,
13670.
24 B. Berlin and P. Kay, Basic color terms: their universality and evolution. University of California Press,
Berkeley & Los Angeles 1969; Kay and Maffi, Color Appearance and the Emergence and Evolution of Basic
Color Lexicons; Haynie and Bowern, Phylogenetic approach to the evolution of color term systems, p. 13667.
25 Haynie and Bowern, Phylogenetic Approach to the Evolution of Color Term Systems, p. 13669.
26 Anna Wierzbicka. Why There Are No ‘Colour Universals’ in Language and Thought, “The Journal of the
Royal Anthropological Institute” (14) 2, (2008), p. 408.
27 Ibidem.
28 Ibidem.
29 Ibidem, p. 417.
30 Ibidem, pp. 408, 419.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
203
The Tibeto-Burman Family
Tibeto-Burman languages are a subgroup within the somewhat contested Sino-Tibetan
family, which broadly includes the massive array of Chinese languages. Taxonomic and
genetic descriptions are rife with complexity and disagreement, with van Driem proposing
“Trans-Himalayan”31 to account for the linguistic geography of the region. Tibeto-Burman
languages stretch from Kashmir to Vietnam, and as a result of massive areal range and
huge internal diversity, genetic affiliations remain hotly contested.32 With this disclaimer,
we nevertheless continue to use Tibeto-Burman to describe this language grouping as it
remains a commonly recognized category at the time of writing. The focus of this article
is on the colour terminology of the Rāī-Kiranti ethnolinguistic subgrouping of Eastern
Nepal, with further data from neighbouring Tibeto-Burman languages to the extent that
such data is available. This article provides a particular emphasis on colour terminology
in Thangmi (UNESCO: definitely endangered, 33,500 speakers, ISO 639-3: thf),33 also
known as ‘Thami’, using data drawn from the primary field work of one of the co-authors.
Thangmi exhibits one of the smallest ranges of colour in the entire family as well as
a grammatical alienability distinction with the colour ‘red’. Each time that we introduce
a new language in the course of this article, the vitality status, number of speakers,
and ISO 639-3 code are also supplied. While census data in the Himalayan region are
notoriously unreliable, and have been known to omit elements of the population on the
fringes of the community or geographically distant and culturally distinct from major
centres of political influence, we have chosen to include official census data if only
to emphasize the important impact that speaker size can have with regard to language
change and maintenance.34 Additional languages highlighted from a variety of sources
with idiosyncratic colour inventories in this typological study include Yakkha (critically
endangered, 14,648 speakers, ISO 639-3: ybh)35 that possesses a ‘red/ non.red’sub-binary
within its own system and Ladakhi (vulnerable, 105,000 speakers, ISO 639-3: cna;lbj;tkk)36
that exhibits a complex and artistic colour schema.
31 George van Driem, Tibeto-Burman Subgroups and Historical Grammar, “Himalayan Linguistics” (10) 1,
(2011), p. 32.
32 David Bradley, Tibeto-Burman Languages and Classification, “Pacific Linguistics”, Canberra 1997, p. 1;
Scott DeLancey, Sino-Tibetan Languages, in: The World’s Major Languages, Oxford University Press, Oxford 1987,
p. 693; Tej R. Kansakar, The Tibeto-Burman Languages of Nepal: A General Survey, Kirtipur 1993, pp. 165–166;
G. Thurgood and R.J. LaPolla (eds.), The Sino-Tibetan Languages (Vol. 3), “Psychology Press”, (2003), p. 7.
33 UNESCO Atlas of the World’s Languages in danger, Viewed 24 July 2017, <http://www.unesco.org/languagesatlas/index.php>.
34 Mark Turin, Time for a True Population Census: the Case of the Miscounted Thangmi, “Nagarik (Citizen)”,
2 (4), Kathmandu (2000), pp. 14–19.
35 UNESCO Atlas of the World’s Languages in danger.
36 Ibidem.
204
MARK TURIN, BENJAMIN CHUNG
Adjectives and (Re)duplication in Tibeto-Burman Languages
Salient to the present discussion, many Tibeto-Burman languages share two noteworthy
characteristics: adjectival classes and the occurrence of (re)duplication. First, in TibetoBurman languages, adjectives are usually derived from verbs, clear examples of which
exist in Thulung Rai (definitely endangered, 14,034 speakers, ISO 639-3: tdh),37 Yakkha,
and Jero (vulnerable, 2,000 speakers, ISO 639-3: jee),38 each which will be covered
in some detail below.39 The derivational nature of adjectives in the family reflects the
general lexicon of Tibeto-Burman languages, which are known for a rich inventory
of complex and descriptive verbs and verb paradigms. Second, reduplication, which
Abbi describes as, “[the] repetition of all or part of a lexical item carrying a semantic
modification”40 does not necessarily entail that a part of a word is in fact reduplicated,
but rather reproduced or simply duplicated in some manner. Abbi describes this process as
a “common phenomenon”41 within this language family, demonstrating a sprachbundian
effect shared between neighbouring Indo-Aryan languages (and some Dravidian languages)
spoken in the region. For this reason, reduplication is not a characteristic exclusive within
the Tibeto-Burman family, but rather an areal feature throughout South Asia. For more
information on reduplication, we direct the reader to classic studies by Emeneau (1956)
and Masica (1976).42
In Tibeto-Burman languages, these two features can appear in concert with one
another, resulting in reduplicating onomatopoeic verbs that are particularly attested in
the Bodic subgroup that includes the Rāī-Kiranti languages.43 For example, in Magar
(definitely endangered, 489,383 speakers, ISO 639-3: mgp, mrd),44 khasak-khusak ka means
‘whisper’45 while phawk phawk jat means ‘pat something’.46 In Magar, onomatopoeic verbs
are a complex class that may combine with nouns as well as full lexical verbs, and can
function as adverbial components.47 In South Asia, reduplicated constructions commonly
Ibidem.
Ibidem.
39 Aimeé Lahaoussois, “Aspects of the Grammar of Thulung Rai: an Endangered Himalayan Language”
(PhD diss., University of California Berkeley, 2002), p. 196; Diana Schackow, A Grammar of Yakkha, Berlin 2015,
p. 80; Jean R. Opgenort, A Grammar of Wambule: Grammar, Lexicon, Texts, and Cultural Survey of a Kiranti
Tribe of Eastern Nepal. Brill, Leiden 2004, p. 115.
40 Anvita Abbi, Reduplication in Tibeto Burman Languages of South Asia, “Southeast Asian Studies” (28) 2,
Kyoto (1990), p. 171.
41 Ibidem, p. 171.
42 We thank one of our anonymous reviewers for emphasising the importance of earlier publications that
highlight reduplication throughout South Asia.
43 Karen Grunow-Harsta, A Descriptive Grammar of Two Magar Dialects of Nepal: Tanahua and Syangja
Magar, The University of Wisconsin-Milwaukee 2008, p. 150.
44 UNESCO Atlas of the World’s Languages in Danger.
45 Ibidem, p. 151.
46 Ibidem.
47 Ibidem, p. 152.
37
38
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
205
appear in adverbs as well as in expressives which exhibit the senses.48 Abbi suggests that
two forms of reduplication – morphological and lexical – should be considered to be areal
features of the region.49 Although reduplication is manifested in a range of lexical items,
not all Tibeto-Burman languages reduplicate as predictably or productively. In Thado,
for example, colour and taste modifiers never reduplicate, whereas colour terms in our
data set are often reduplicated.50 As is to be expected in a language family of such size,
areal reach and internal complexity, it is difficult to discern specific characteristics that
may be said to be shared by or common to all Tibeto-Burman languages.
Loans in Tibeto-Burman Languages
Many Tibeto-Burman languages, and certainly those spoken in Nepal, borrow from
socially and politically dominant languages such as Nepali. In at least two cases outlined
below, the colour term for ‘yellow’ has been taken from the Nepali word for turmeric
(besār), denoting the rich colour of the processed spice, Curcuma longa.51 Similarly,
Gyarong (Ethnologue: vigorous, 83,000 speakers, ISO 639-3: jya),52 a Tibeto-Burman
language spoken in Sichuan, China, exhibits colour loans from the more dominant languages
of the region, Sichuanese Mandarin and Tibetan.53 This dynamic of lexical borrowing and
ultimately language shift are crucial for understanding the rate of endangerment among
Indigenous languages across the globe, not just those within the Tibeto-Burman family.
For this reason, and throughout this study, we highlight these loan words when we have
been able to identify them and are confident about their provenance.
Grzega explores why languages borrow words and identifies a number of causes
that may contribute to our understanding of borrowed of colour terms in Tibeto-Burman
languages. Explanations for borrowing that are relevant to our research include, among
others: “feeling of insufficiently differentiated conceptual fields”;54 “rise of a specific
conceptual field”;55 “political or cultural dominion of one people by another”;56 “mere
Abbi, Reduplication in Tibeto Burman Languages, p. 172.
Ibidem, p. 171.
50 Ibidem, p. 177.
51 Gerard J. Tolsma, A Grammar of Kulung, Languages of the Great Himalayan Region, Brill, Leiden 2006,
p. 40; Schackow, A Grammar of Yakkha, p. 162.
52 David M. Eberhard, Gary F. Simons, and Charles D. Fennig (eds.). 2015. Ethnologue: Languages of the
World. Twenty-second edition. Dallas, Texas: SIL International.
53 Yasuhiko Nagano, A Preliminary Note to the Gyarong Color Terms, Osaka (2008), p. 101.
54 Uriel Weinreich, Languages in Contact: Findings and Problems (1), The Hague (1953), p. 59; as cited by
J. Grzega, Borrowing as a Word-Finding Process in Cognitive Historical Onomasiology, “Onomasiology Online”
(4) (2003), p. 23.
55 Joachim Grzega, Some Aspects of Modern Diachronic Onomasiology, (2002), p. 1030; as cited by Grzega,
Borrowing as a Word-Finding Process, p. 23.
56 Gerd Fritz, Change of Meaning and Change of Vocabulary, “Sociolinguistics: An International Handbook of
the Science of Language and Society” (11), (1988), p. 1622; as cited by Grzega, Borrowing as a Word-Finding
Process, p. 23.
48
49
206
MARK TURIN, BENJAMIN CHUNG
oversight or temporary lack of remembering the indigenous name”;57 and “low frequency
of indigenous words and instability of words within a region.”58 At this juncture, it is
essential to acknowledge that the presence of loan words and borrowings in a language
do not make these additions any less legitimate as items in the lexicon. Rather, the
inclusion of lexemes (whether they be calques or loans) in these languages are the end
result of complex, historical and often intersecting factors.
In the context of the Tibeto-Burman languages spoken in Nepal, the most salient
motive for borrowing from the above list is arguably “political or cultural dominion of
one people by another.”59 UNESCO considers most of the languages that we cover in this
airtlce to be endangered or vulnerable, a compelling indication of the huge socio-political
forces that threaten Indigenous languages and cultures in the region. It is entirely credible
to propose that such rapid transformations – through rampant urbanisation, a strengthening
media sector, social and political upheavals and compulsory primary education – could
also further catalyze, “[a] feeling of insufficiently differentiated conceptual fields”60 or
the, “[subsequent] rise of a specific conceptual field”61 that may have been previously
absent in these languages, thus stimulating innovation and changes in the lexicon.
When it comes to Indigenous terms and lexicon, the causes behind borrowing must be
carefully scrutinized. As Grzega notes, “what is a low frequency rate of a word? Does it
mean that the concept is rarely talked of? Does this then include that infrequent concepts
have a tendency to be named with a loanword?”62 Wierzbicka draws our attention to
the prevalent if dangerous tendency of evaluating and comparing Indigenous names and
words to the lexicon of world languages (especially English), all of which can lead to
skewed and highly partial analysis.63 When discussing loans, it is therefore important
to bear in mind Wierzbicka’s wise counsel: we must challenge ourselves to distinguish
between instances when there is an actual “lexical gap”64 and situations which can be
explained as an alternative viewpoint or worldview that cannot be easily lexicalized or
rationalized in another language. It is also necessary to recognize the presence of – and
differences in – metaphoric and literal meaning that may arise in these terms in their
respective languages. Complex uses cannot be easily determined by outside researchers
57 Weinreich, Languages in Contact, p. 60; U.G. Baranow, “Studien zum deutsch-portugiesischen Sprachkontakt
in Brasilien” (Master’s diss., Ludwig Maximilian-Universität, München 1973), p. 138; G. Tesch, Linguale Interferenz:
Theoret., Terminolog. u. Method. Grundfragen zu ihrer Erforschung, Tübingen 1978, p. 209, 214; as cited by
Grzega, Borrowing as a Word-Finding Process, p. 24.
58 Weinreich, Languages in Contact, p. 57; M. Scheler, Der Englische Wortschatz: Grundlagen der Anglistik
und Americianistik, Berlin 1977, p. 88; as cited by Grzega, Borrowing as a Word-Finding Process, p. 24.
59 Fritz, Change of Meaning and Change of Vocabulary, p. 1622; as cited by Grzega, Borrowing as a WordFinding Process, p. 23.
60 Weinreich, Languages in Contact, p. 59; as cited by Grzega, Borrowing as a Word-Finding Process, p. 24.
61 Grzega, Aspects of Modern Diachronic Onomasiology, p. 1030; as cited by Grzega, Borrowing as a WordFinding Process, p. 23.
62 Grzega, Borrowing as a Word-Finding Process, p. 24.
63 Wierzbicka, Why There Are No ‘Colour Universals, p. 417.
64 Ibidem, p. 408.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
207
and require further semantic and pragmatic analysis. In addition, we acknowledge that
historical sound changes within the Tibeto-Burman family make it difficult to accurately
identify and describe all cognates at this time. To that end, in this article, we take no
position on the cause or reasons for specific borrowings, and restrict ourselves to indicating
cognates and loans that are of particular note or interest in the present discussion.
Basic Tibeto-Burman Colour Terminology in Perspective
Across the Tibeto-Burman family, we have found no indication of an average number
of colour terms that a language might exhibit. Languages like Gyarong boast approximately
eleven distinct colour terms, while others such as Thangmi have only three.
For example, Kham (definitely endangered, 30,000, ISO 639-3: kgj, kip)65 attests
only 4 colour terms and thus would be considered a Stage III language within Berlin &
Kay’s initial theory.66
Table 1. Kham Colours67
Kham
English
‘molo
black
palo or ‘pəlã:
white
‘gyahm-nya
red
‘pĩ:-nya
green
It is of note there are various dialects of Kham, and the terms provided derive from
the Takale dialect, which is the most prestigious dialect for the Western Parbate subgroup.68
Watters argues that there are visible relationships with other proximous languages including
Chepang and Thakali (which will be discussed in this paper); although he concedes that
his position is speculative.69
In Kham, ‘white’ and ‘black’ do not end with <-nya>, a morpheme present in other
adjectives including ‘red’ and ‘green’. This distinction is of interest as ‘black’ and ‘white’
represent the basis for human colour perception. Nevertheless, Kham can be positioned as
a typical language in the Berlin & Kay scale, lexicalizing four colours including ‘green’
but not any form of ‘yellow’.
65
UNESCO Atlas of the World’s Languages in danger.
D.E. Watters, A Grammar of Kham, Cambridge 2002, pp. 452–453.
67 Ibidem.
68 Ibidem, pp. 12, 432.
69 Ibidem, p. 432: Watters lists a variety of sources from which he has gathered Kham lexicon over the years
in this section of his grammar, but does not specify an exact reference for each lexeme in the subsequent text.
Please refer to the original grammar for these sources.
66
208
MARK TURIN, BENJAMIN CHUNG
By contrast, Kulung (vulnerable, 18,686 speakers, ISO 639-3: kle),70 an Eastern Rāī
language, possesses five distinct Indigenous colours and one loan: besarwa ‘yellow’,
derived from the Nepali word for ‘turmeric’ besār.71 Interestingly, ‘yellow’ is distinct
from ‘light yellow’, which is encompassed within the Kulung term for ‘white’: omlo:pa.
Moreover, reduplication is observed for most terms other than ‘yellow’ and ‘light yellow’.
In Kulung, these adjectives are adjectival in nature and are not derived from verbs, setting
them apart from other Tibeto-Burman relatives.
Table 2. Kulung Colours72
Kulung
1
English
omlo:pa
white, light yellow
gugurpa
black
halala:pa (possibly Nep.)
red, pink
gigippa
green, blue, purple
besarwa (Nep.)
yellow
momoppa
brown
am-pʰaji
gugur-yo
cʰuː-a
kʰat-a.
yours-rucksack black-INT
be-PT
go-PT
Yours rucksack has become pitch-black.73
In the above example, the Kulung word for ‘black’ gurgurpa is shown with an
intensifier suffix <-yo> instead of –pa, which would be typical for unmodified colours
and other adjectives.74 In this case, ‘black’ is not simply ‘black’ but ‘very black’ or
‘pitch-black’.75 It is not clear how besarwa might intensify as it lacks the –pa suffix
typical of colours. It is also noteworthy that besarwa is not a native word in the Kulung
language, but a loan from Nepali in which it is not a colour per se but rather the name
of a spice.
In Kulung, as in other Tibeto-Burman languages, including Thangmi, colours can be
nominalised to describe people and things as, “the one who is …”76
70
71
72
73
74
75
76
UNESCO Atlas of the World’s Languages in Danger.
Tolsma, Grammar of Kulung, pp. 1, 40.
Ibidem, p. 40.
Ibidem.
Ibidem, p. 41.
Ibidem.
Ibidem, p. 102.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
209
omlo:pa-kə
white-SUB
the white one77
2
In the above example, the subordinating suffix <-kə> attaches to omlo:pa to describe
someone or something that is ‘white’. In principle, omlo:pa-kə could also mean ‘the
light yellow one’.
Thulung Rai (definitely endangered, 14,034 speakers, ISO 639-3: tdh)78 shares some
features with Kulung, including some recognizable cognates. Thulung Rai does exhibit
some discrete colours that are not attested in Kulung, and the Thulung colour inventory
is derived from verbs.79
Table 3. Thulung Rai Colours80
Thulung Rai
English
lalam
red (possibly from Nepali lāl ‘red’)
gigim
green
kekem
black (possibly from Nepali kālo ‘black’)
ʔoʔom
yellow
bubum
white
nunum
blue, ‘green-blue’ (possibly from Nepali nilo ‘blue’)
Interestingly, nunum is described as both ‘blue’ and ‘green-blue’ by Lahaussois.81 As
noted above, this discrepancy may be down to different perceptions of colour boundaries
and does not necessarily imply that there is no term for ‘blue’ in the language, but rather
that some ‘greens’ may be incorporated into the category ‘blue’. Other languages in
the Tibeto-Burman family, such as Gyarong, exhibit similar tendencies with the colours
‘blue’ and ‘green’, most likely indicative of the natural simultaneous perception of ‘blue’
and ‘green’.82
Limbu (definitely endangered, 300,000 speakers, ISO 639-3: lif)83 possesses a smaller
range of colours than Kulung and Thulung Rai, with a distinction made between the
Ibidem.
UNESCO Atlas of the World’s Languages in Danger.
79 Tolsma, Grammar of Kulung, p. 40; Lahaoussois, “Aspects of the Grammar of Thulung Rai,” pp. 196, 375.
80 Lahaoussois, “Aspects of the Grammar of Thulung Rai,” pp. 196, 375.
81 Ibidem.
82 L.M. Hurvich and Dorothea Jameson (trans.), Outlines of a Theory of the Light Sense, Cambridge, MA 1964;
as cited in Kay and McDaniel, Linguistic Significance of the Meanings of Basic Color Terms, p. 620.
83 UNESCO Atlas of the World’s Languages in Danger.
77
78
210
MARK TURIN, BENJAMIN CHUNG
colour itself and the quality of that colour, with affixes determining such variation.84
Grammars of Limbu prepared by Weidert & Subha and van Driem show pronounced
phonological and semantic differences.85
Van Driem categorizes only four colours as cardinal colours: mak ‘black’, bhɔ ‘white’,
hɛt ‘red’, and hik ‘green’, and emphasizes the uniqueness of this colour set as a result of
their freeness and ability to change depending on their affixes.86 For example, ɔmdaŋba
‘yellow’ is only grammatical with a certain affix <-taŋba> and has a limited range of
colour that it represents.87 For these reasons, van Driem does not consider ‘yellow’ to
be a main colour in Limbu, although it is certainly present.
3
a-him-min
kubhɔra coˑk.
kɛ-him-min
my-house-ABS
white be.
yours-house-ABS
My house is white. Your house is all red.88
kuhɛtla-kuhɛtla
red-red
coˑk.
be
In their grammar, Weidert & Subha reference a different root for ‘green’ sɔre and
offer hiˑk as ‘yellow’.89 In addition, ‘to be blue’ is documented as either phiˑŋ-lɔˀma or
kubhiˑŋla, where these terms do not appear in van Driem’s work.90 Aside from these
differences, Weidert & Subha’s terms for ‘black’, ‘white’, and ‘red’ are equivalent to
those provided by van Driem.
The following tables provides examples from both grammars to demonstrate affixation.
Table 4. Limbu Colour Affixation91
Affix
Meaning
Examples
-lɔˀma/ -yaˑpma
*lɔʔmaʔ
(van Driem 1987:23)
to be –
*to appear
hɛˑt-lɔˀma ‘to be red’ (251)
hiˑk-lɔˀma ‘to be yellow’ (252)
-kɛlɔˀba
pure adjectival form
phiˑŋ-lɔˀma ‘blue’
(Weidert & Subha 1985:323)
-taŋba
‘that which is, he who is’
hiktaŋba ‘green one’ (van Driem
1987:424)
84 A. Weidert and B. Subha, Concise Limbu Grammar and Dictionary: Concise Limbu Grammar, Nominal
Paradigms and Verbal Paradigms, Concise Limbu-English Dictionary, English-Limbu Vocabulary, 1985, p. 52.
85 Weidert and Subha, Concise Limbu Grammar and Dictionary, p. 52; George van Driem, A Grammar of
Limbu, Walter de Gruyter, Berlin 1987, pp. 23–24.
86 Van Driem, Grammar of Limbu, pp. 23, 25.
87 Ibidem, p. 25.
88 Ibidem, p. 24.
89 Weidert and Subha, Concise Limbu Grammar and Dictionary, pp. 385, 409.
90 Ibidem, p. 374.
91 Weidert and Subha, Concise Limbu Grammar and Dictionary, pp. 52, 251–252, 286, 323, 336; van Driem,
Grammar of Limbu, pp. 23, 424.
211
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
Affix
Meaning
-yɔˀyɔˀba
Examples
-ish
sɔreˑˀyɔˀyɔˀba ‘greenish’ (Weidert
& Subha 1985:336)
mak-yɔˀyɔˀba ‘blackish’ (Weidert
& Subha 1985:286)
There are at least four recognizable dialects of Limbu, with variable pronunciation
between them.92 These differences appear mostly as predictable sound changes within
affixes and may contribute to some of the variations attested in these two grammars.
In Limbu as in Kulung, colour can be used to describe something or someone.
Applying the suffix <-taŋba>, colours can either behave like adjectives or nominalise
fully to become nouns. In the following example, it is apparent that such terms can be
used metaphorically, in this case describing a European person as ‘white eyed’.
4
mikphuʔla meˑn
laʔba,
ku-mik
phɔ-daŋma.
European
NOT
perhaps,
her-eye
white-vālā/f
Maybe she’s not European (a white-eye), but she sure is white-eyed!
(i.e. she sure does look like one).93
While adjectives in Sunwar (vulnerable, 26,611 speakers, ISO 639-3: suz)94 are mostly
verbal nouns, colour terms do not belong to this class. Other adjectives that do not derive
from verbs include loanwords from Nepali.95
Table 5. Sunwar Colour Terms96
Sunwar
English
gīk
light green, light blue
nilo (from Nepali)
dark blue
buʃ
white
kher
black
lal (from Nepali)
red
92 Weidert and Subha, Concise Limbu Grammar and Dictionary, pp. 4–7; van Driem, Introduction to Grammar
of Limbu, by van Driem, p. xxii.
93 Van Driem, Grammar of Limbu, p. 24.
94 UNESCO Atlas of the World’s Languages in Danger.
95 D. Borchers, A Grammar of Sunwar: Descriptive Grammar, Paradigms, Texts and Glossary, Brill, Leiden
2008, p. 93.
96 Ibidem.
212
MARK TURIN, BENJAMIN CHUNG
Overall, Sunwar presents a challenge to the traditional Berlin & Kay model as there are
approximately five colour terms attested in the language, including two terms for ‘blue’.
Additionally, Sunwar boasts two additional adjectives that pertain to colour: ‘colourful’
jirjir and ‘brilliant’ ojela.97
5
ā
ke
buʃ
buʃ
cā
tam.te.me
buʃ
she/he
POSS white
white hair
see.PT-3p.3p/svi
white
tam.te.me.
see.PT-3p.3p/svi
They saw her white hair and her white skin.
(Excerpt from A foreigner in Bhujī by Śobhā Mulicā Sunuvār)98
ʃe
flesh
In Sunwar, one term for ‘blue’ is a clear loan from Nepali nilo and covers ‘dark
blue’, while the other term, gīk, is a native term that encompasses a spectrum representing
‘light green’ to ‘light blue’.99 According to the Berlin and Kay model, only Stage V
languages possess a word for ‘blue’, along with colour terms for ‘green’ and ‘yellow’. In
Sunwar, ‘blue’ is incorporated into the term for ‘green’, and there is no discrete lexical
distinction between the two colours nor is there an apparent and specific term for ‘yellow’.
Additionally, since Borchers specifies a distinction between ‘dark’ and ‘light’ variants, this
introduces the question of which of the Sunwar colour terms is closer to ‘true blue’ or
rather, as Kay & McDaniel would have it, which has the highest membership to ‘blue’.100
Another visible cognate is lal for ‘red’, similar to terms in Kulung, halala:pa,101 and
Thulung Rai, lalam.102 In Nepali and Hindi, lāl is one of the terms for ‘red’, and for
this reason, it is highly likely that lal is a direct loan from a neighbouring Indo-Aryan
language.
Spoken in southeastern Nepal, Dhimal (severely endangered, 20,000 speakers, ISO
639-3: dhi)103 is divided into two dialects.104 Like many of their linguistic cognates and
cousins, adjectives in Dhimal are modified from verbs with the addition of a morpheme:
<–ka>.105 Generally, the morpheme <–ka> does not appear with borrowed adjectives
with the verbal forms of these colours typically ending with the morpheme: <-li>.106
97
98
99
100
101
102
103
104
105
106
Ibidem.
Ibidem, pp. 261, 265.
Ibidem, p. 93.
Ibidem, p. 624.
Tolsma, Grammar of Kulung, p. 40.
Lahaussois, “Aspects of the Grammar of Thulung Rai,” p. 196.
UNESCO Atlas of the World’s Languages in Danger.
J.T. King, A Grammar of Dhimal, Languages of the Greater Himalayan Region, Brill, Leiden 2009, pp. 1–2.
Ibidem, p. 52.
Ibidem, pp. 511, 533, 536, 605.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
213
Table 6. Native Dhimal Adjectives107
Dhimal
English
da:ka
black
yauka
yellow
i:ka
red
je:ka
white
The four basic colours shown in the above table situate Dhimal as a Stage III language
and tend to represent spectrums for Dhimal speakers. As King details, “‘black’ covers
the range from black and dark brown, to dark purple and blue, and the term i:ka ‘red’
ranges from red to reddish brown.”108 Je:ka can also mean ‘Caucasian’.109
6
da:-ka
mundha
black-NOM
stump
blackened stump110
These base colours also appear in a variety of other terms, particularly in terminology
relating to local fauna.
Table 7. Animal Terms in Dhimal111
Dhimal
English
da:ka kawa
large-billed crow, Corvus macrorhynchos
da:ka koʔsa
black monitor lizard
i:ka koʔsa
red monitor lizard
i:ka nhõya
rhesus monkey, Macaca mulatta
i:ka nhamui
small red ant
yauka koʔsa
yellow monitor lizard
In addition, other Dhimal terms incorporate these colours, including terminology used
to describe traditional dress and references to skin tone.
107
108
109
110
111
Ibidem,
Ibidem.
Ibidem,
Ibidem,
Ibidem,
p. 54.
p. 536.
p. 52.
pp. 511, 533, 536, 605.
214
MARK TURIN, BENJAMIN CHUNG
Table 8. Miscellaneous Colour-derived Terms in Dhimal112
Dhimal
English
da:ka beraŋ
nickname for a dark-skinned woman
da:boʔna
black boʔna (traditional wrap skirt) with two red
stripes around the waist
da:kiculhoʔka
black as night, pitch black
je:pa
fair, light-coloured
je:pa jeŋka
fair-skinned
‘Orange’ i:tatarpa and ‘pink’ i:lalhaipa are also attested in the language, both of which
are derived from the basic i:ka ‘red’.113 Interestingly, these colours use the morpheme
<-pa> instead of <-ka>, a process more typically observed for adverbials.114
Adjectives in Dhimal can undergo reduplicative processes to express different aspects
and intensities, such the diminutive, which is produced using the morpheme: <-co:>.115
In particular, King suggests that the morpheme <-co:> is a cognate with the Limbu
‘be small’ cuk-maʔ.116 When applied to colour terms, the addition of <-co:> changes
a description from ‘red’ i:ka to ‘reddish’ i:co:co:ka.117 Another term that can reduplicate
is yauka, which intensifies ‘yellow’ to a ‘deep golden yellow’, yauyauka.118
While the closely related Jero and Wambule languages (vulnerable, 4,471 speakers,
ISO 639-3: wme)119 share the most similarities in colour range with each other, there
are some striking differences in their respective lexicons.120 Both languages have
white, black and red, and additionally yellow and green, which would theoretically
place them at stage VI along Berlin and Kay’s categorical scale, although Wambule
possesses more terms than Jero even if not all of them are native to the Wambule
language.
Ibidem, pp. 511, 536.
Ibidem, p. 54.
114 Ibidem.
115 Ibidem.
116 Ibidem.
117 Ibidem.
118 Ibidem, p. 605.
119 UNESCO Atlas of the World’s Languages in Danger.
120 Jean R. Opgenort, A Grammar of Jero: With a Historical Comparative Study of the Kiranti Languages,
Languages of the Greater Himalayan Region, Brill, Leiden 2005, p. 5.
112
113
215
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
Table 9. Jero and Wambule Colours121
Jero
Wambule
Colour
bupcip ~ bubjɛŋmo ~ bubu (Āmboṭe dialect)
bubu, bu (n) ~ bubjwam (adj)
white
khucɛm (Mohanṭāre dialect) ~ khucɛp (Āmboṭe
dialect) ~ khucɛŋmo (Āmboṭe dialect)
khuce, khuɖ (n) ~ khuc(c)yam,
khuccyaŋmo (adj)
black
laka ~ lacip (Āmboṭe dialect)
laka, lak (n) ~ lakajwam, lakajim red
(adj)
pʌhẽlo (Āmboṭe dialect) (from Nepali for
‘yellow’ pahẽlo) ~ waʔɔmjimo (Mohanṭāre
dialect) ~ waʔɔmjɔkto (Mohanṭāre dialect)
waʔwam (n) ~ waʔwamjwam,
waʔwamjim (adj) ~ wamcam
‘to be yellow’ (v)
hʌriyo (Āmboṭe dialect) from Nepali for ‘green’; hʌriyo (Nep.); sisi
pʌlʌi ‘becoming green’ (specifically with plants)
(Mohanṭāre dialect); pʌlʌi dumcam ‘become
green’ (Mohanṭāre dialect)
yellow
green
The distribution of colour terms in Jero is uneven. For some colours like ‘red’ and
‘white’, intensity can be increased through reduplication, i.e. laka-laka ‘very red’
and bubuyaya ‘very or purely white’ (Āmboṭe dialect).122 Opgenort does not volunteer
the same approach to explain the intensification of ‘black’, although it is theoretically
possible. It is important to note that only the Āmboṭe dialect of Jero seems to exhibit
this feature.
Wambule does have terms for ‘intense black’: khuce-khuɖ ‘very black’ and khuccekhuɖ ‘extremely black’.123 ‘Very red’ also exists, with the final a optionally omitted from
laka-laka, as in laka-lak,124 and another term for ‘white’ is also documented. Although
Opgenort does not describe an additional term for ‘very white’ in Wambule in the manner
that he does for Jero, he does note that there are specific terms for mixed colours in
Wambule.
Table 10. Additional Wambule Colour Terms125
Wambule
Term
Any colour mixed with white
blwaŋce
Any mixed colour
blwabjwam
121 Opgenort, A Grammar of Wambule, pp. 227–228, 613, 658, 762, 797, 830, 857, 860; Opgenort, A Grammar
of Jero, pp. 116, 283, 294, 307, 320–321.
122 Opgenort, A Grammar of Jero, pp. 207, 320.
123 Opgenort, A Grammar of Wambule, p. 762.
124 Ibidem, p. 830.
125 Ibidem, p. 857.
216
MARK TURIN, BENJAMIN CHUNG
In addition, Wambule possesses words for ‘purple’ waʔwal (noun) and the Nepali loan
nilo for ‘blue’ (adj.) that Jero does not, theoretically placing Wambule higher than Jero in
the Berlin & Kay hierarchy. The existence of blwaŋce and blwabjwam in Wambule raises
further questions of how colour terms that describe colour mixture might be incorporated
into the existing theory.
Interesting parallels exist between the implementation of colour terminology in the
lexicons of Thangmi and Yakkha. Despite the constrained range of lexical items to express
colour in Thangmi, the colour terms that do exist have a relatively wide range of uses.
In Yakkha, by contrast, while a wide array of colour terms are attested, all other colour
terms are secondary to the categories of ‘black’, ‘white’ and ‘red’, which correlate to
Thangmi’s limited inventory.
Table 11. Yakkha Colours126
Yakkha
English
phamna
red
phimna
green, blue, (yellow)
phuna
white
makhurna
black
phalik-phalikna
reddish/pink/violet (various light and dark shades)
phiʔlik-phiʔliŋna
greenish, blueish (sky blue, petrol, light green)
phiriryaŋna
yellow (specifically food)
besareʔna (loan from Nepali, as in Kulung) yellow
phutiŋgirik
bright white
phutlek-phutlekna
light grey, light yellow, light pink, beige
Yakkha possesses at least 11 terms for colours, many of which represent fuzzy sets
as understood by Kay & McDaniel in their extension of Berlin & Kay’s original theory.
Nevertheless, there are major distinctions between ‘black’ and ‘white’, makhurna and
phuna, and ‘red’ and ‘non.red’, phamna and phimna.127
7
paŋ=be
phu=ha=huŋ
makhur=ha
caleppa,
house=LOC
white=NMLZ.NC=COM
black=NMLZ.NC
bread,
macchi
khicalek=nuŋ
cuwa
py-a.
pickles,
rice_dish=COM
beer
give-PST[1.P]
At home, they gave us white and black bread, pickles, khichadi and beer.128
126
127
128
Schackow, A Grammar of Yakkha, p. 163.
Ibidem, p. 162.
Ibidem, p. 155.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
8
sumphak
loʔa=na
leaf
like=NMLZ.SG
as green as a leaf129
217
phim=na
non-red=NMLZ.SG
In example 8 above, the difference between ‘non.red’ and ‘red’ is highlighted. Although
the word for ‘green’ does not even appear in the sentence, the use of phimna ‘non.red’
to characterize a leaf as typically ‘not red’ allows the sentence to be understandable to
Yakkha speakers as ‘green’.
Another example of the same feature is provided below in example 10:
9
massi
loʔa=na
ink
like=NMLZ.SG
as blue as ink130
phim=na
non-red=NMLZ.SG
With massi (a loan from Nepali) in place of sumphak ‘leaf’, the sentence becomes
about the colour ‘blue’ while still utilizing the ‘non.red’ identifier of phimna.
In Yakkha, several additional lexemes modify colours, such as om(na) ‘bright, light’,
kuyum(na) ‘dark’ and chyaŋchyaŋ(na) ‘transparent’.131
10 wa=ci
ŋ-ga-ya=hoŋ=go
om
chicken=NSG 3PL-speak-PST=SEQ=TOP bright
As the cocks crowed, it had already dawned.132
keks-a-khy-a.
become-pst-V2.go-pst[3sg]
In example 10 above, om ‘bright’ is used metaphorically to convey the sense of
‘dawn’, an extension of meaning that is also attested in Thangmi.
In Yakkha, it is phonologically noteworthy that all colour terms have an initial bilabial
plosive {[b], [p], [m]} and similarly interesting that besareʔna (a loan from Nepali besār
meaning ‘turmeric’) is also attested, perhaps an indication of the importance of trade
with Nepali speakers from whom turmeric would have been acquired.
In contrast to Yakkha’s large and impressive inventory, the Thangmi language has
adjectives to express only three, distinct colours: ‘black’, ‘white’ and ‘red’ (all Thangmi
data provided by co-author Turin). While older speakers insist that there were once terms
for a greater range of colours on the spectrum, there is no persuasive evidence of this.
According to Berlin & Kay’s aforementioned proposed categories of colour terminology,
as presented in their Basic Colour Terms, Thangmi would be an example of a typical
Stage II language, with Indigenous lexical items for ‘black’, ‘white’ and ‘red’ only. All
other colour terms are borrowed from Nepali. Alongside terms for these three primary
colours, Thangmi has native adjectives meaning ‘dark’ and ‘light’, but these cannot
129
130
131
132
Ibidem, p. 162.
Ibidem.
Ibidem, p. 163.
Ibidem, p. 544.
218
MARK TURIN, BENJAMIN CHUNG
be used to modify the intensity of a colour. Thangmi colour terms and the associated
adjectives expressing lightness and darkness from the Dolakhā dialect are presented in
Table 12 below.
Table 12. Thangmi Colour Adjectives133
Thangmi
English
ḍiŋ-ḍiŋ
red
kiji
black
ubo
white
athaŋ
light (as in brightness)
ukhiŋ
dark
In Thangmi, alongside the standard meaning of ubo ‘white’, as illustrated by
example 11, ubo ‘white’ can precede mi ‘person’ to render ubo mi (white person)
‘foreigner, white-skinned person’ similar to the example previously provided for Limbu,
as in examples 12 to 15 below.
11 găi-go
miŋ
ubo
hok-Ø-du,
tara
aye
I-GEN
cloth
white
be-sAS-NPT
but
much
bu-si-ta-ŋa-le
kiji
thah-Ø-an.
cover-REF-IPP-1s-PCL
black
be-sAS-3S/PT
My clothes are white, but because I have worn them for so long, they
have become black.134
12 ubo
mi-ko
camăica-pali-ye
oste-ko
huca-kăi
white
person-GEN
woman-p-ERG
self-GEN
child-PM
cyocyo
ma-pi
isiy-eŋ-du.
breast
NEG-give
say-pAS-NPT
They say that white women don’t give their children the breast.135
13 ubo
mi-ko
pepelek
aye
white
person-GEN
money
much
White people have a lot of money.136
hok-Ø-du.
be-sAS-NPT
M. Turin, A Grammar of the Thangmi Language (2 vols): With an Ethnolinguistic Introduction to the Speakers
and Their Culture, Languages of the Greater Himalayan Region, Brill, Leiden 2011, p. 314.
134 Ibidem, p. 315.
135 Ibidem.
136 Ibidem.
133
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
219
14 to
isyaŋ
ubo
mi-pali
demca
ra-to-le
that period
white
person-p load
bring-TPP-PCL
ray-eŋ-an.
come.from.level-pAS-3S/PT
That same day some foreigners [lit. white people] came carrying their
own packs.137
15 ‘bore
yo-sa-kăi
ra-ŋa-n’
to
ubo
marriage
look.at-INF-PM
come.from.level-1s-PT
that
white
mi-ye
ŋa-Ø-ŋa-n.
person-ERG
say-sAS-1s-PT
‘I have come to see the wedding’, the white man said to me.138
The adjective ubo ‘white’ can also be combined with the noun bajareŋ ‘local tobacco’,
giving ubo bajareŋ ‘cigarette’ (lit. ‘white tobacco’, as distinct from home-grown tobacco),
as in example 16.
16 to
mi-ye
di-gore
ucyaca
menca-yiŋ
ubo
bajareŋ
that
person-ERG
one-CLF
small
bag-ABL
white
tobacco
kăiy-Ø-u-no,
gă-ye
yo-le
nah-u-n-uŋ.
take.out-sAS-3P-3→/PT
I-ERG
look.at-PCL
put-3P-1s→-1s→/PT
That person took some white tobacco out of a small pouch, and I was
watching.139
The Thangmi adjective ubo can also mean ‘clean’, in which case it is often contrasted
with kiji ‘black’, the latter then meaning ‘dirty’, as in example 17 below. In a South Asian
linguistic context, combining the meanings of ‘white’ and ‘clean’ on the one hand, and
‘black’ and ‘dirty’ on the other, is by no means unusual. In Hindi, for example, safed
‘white’ is etymologically related to sāf ‘clean, fair, bright’.140
17 di
uni
jhari yuw-Ø-an,
to
phow-Ø-an,
one
day
rain
come.from.above-sAS-3S/PT that
be.wet-sAS-3S/PT
pho-Ø-ta-le
to-ko
kiji
maŋ
ubo
thah-Ø-an.
be.wet-sAS-IPP-PCL that-GEN
black
body white
be-sAS-3S/PT
One day it rained and he was totally drenched, and being so drenched,
his dirty body became clean.141
137
138
139
140
141
Ibidem.
Ibidem.
Ibidem, p. 316.
Ibidem.
Ibidem.
220
MARK TURIN, BENJAMIN CHUNG
Alongside the standard use of kiji to mean ‘black’, as in examples 18 and 19, kiji
‘black’ can precede mi ‘person’ to render kiji mi (black person) ‘southerner, plainsman,
Indian’, as in example 20.
18 to-ko
mus
găi-go
uniŋ
that-GEN hair
I-GEN like
His hair is black like mine.142
kiji
black
hok-Ø-du.
be-sAS-NPT
19 to
kiji
semni-ko
bore
kityaŋ
that
black
Tamang-GEN
marriage
three.days.ago
That black-faced Tamang got married three days ago.
thah-Ø-an.
be-sAS-3S/PT
20 ni-ko
kucu-pali-ye
kiji
mi
niy-eŋ-to-le
we-GEN
dog-p-ERG
black
person
see-pAS-TPP-PCL
aghyoy-eŋ-du.
bark-pAS-NPT
When our dogs see dark people, they bark a lot.143
aṭṭhe
very
The adjective kiji ‘black’ can also be used as a proper noun. Slightly dark-skinned
Thangmi children are often called kiji ‘Blackie’, either in their official papers, if they have
any, or as a village nickname. One of the narrative texts recorded concerned a person
named ‘Blackie’, as illustrated by example 21. Unlike Nepali, in which black hounds
and dark male children may be called kāle ‘Blackie (MASC)’, while dark female dogs
and girls are referred to as kālī ‘Blackie (FEM)’, there is no gender differentiation for
‘Blackie’ in Thangmi, and boys as well as girls may be named kiji. Thangmi individuals
may carry the nickname kiji ‘Blackie’ with them into adulthood.
21 di-ka
kiji
name
tha-Ø-du
camăica
one-HNC
blackie
name
be-sAS-NPT
woman
hok-Ø-thyo.
be-sAS3sCOND
There once lived a woman by the name of Blackie.144
Just as ubo ‘white’ is used to mean ‘clean’, so too kiji ‘black’ can mean ‘dirty’ or
‘dark’, as in examples 22 and 23.
22 naŋ-ko
khen
aṭṭhe
kiji
thah-Ø-an.
you-GEN face
very
black
be-sAS-3S/PT
Your face has become very dirty [black].145
142
143
144
145
Ibidem.
Ibidem.
Ibidem, p. 317.
Ibidem.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
221
23 nany-e
thapu
thi-to-le
lak
kiji
thah-Ø-an.
you-ERG
fireplace
touch-TPP-PCL
hand
black
be-sAS-3S/PT
You touched the fireplace and your hand has become dirty [black].146
The final use of kiji ‘black’ is as an intensifier for chokchok ‘darkness’ in the phrase
kiji chokchok (black darkness), best translated as ‘complete darkness’, and illustrated by
example 24.
24 di
uni
ṭhoṇi-ye
kiji
chokchok-te
luma
one
day
old.woman-ERG black
darkness-LOC
partially.husked.rice
kăi-sa
ci-loŋ-Ø-u-no.
remove-INF CAUS-do-sAS-3P-3→/PT
One day the old woman made her remove all the partially-husked rice
in complete darkness.147
In the Dolakhā dialect of Thangmi, the adjective ḍiŋ-ḍiŋ ‘red’ is used both to describe
things that are permanently red, such as a cockerel’s comb (examples 25 and 26) and
more temporary reds, such as flushed cheeks (example 27).
25 nem
thil-sa
beryaŋ,
hyawasa-ŋaŋ
ubo
nasak,
house paint-INF
that.time
upper.part-inside
white
earth
nhawasa-ŋaŋ
ḍiŋ-ḍiŋ nasak-e
thil-eŋ-du.
lower.part-inside
red
earth-INS
paint-pAS-NPT
When it’s time to paint a house, they paint the upper with white earth
and the lower with red.148
26 gare-ko
jire
ḍiŋ-ḍiŋ
tha-Ø-du.
rooster-GEN
crest
red
be-sAS-NPT
The crest of the cockerel is red.[Cockerels’ crests are red] [A cockerel’s
crest is red].149
27 marci
cya-Ø-ta-le
cile
ḍiŋ-ḍiŋ
hot.chilli
eat-sAS-IPP-PCL
tongue
red
150
If you eat chillies your tongue will go red.
tha-Ø-du.
be-sAS-NPT
The adjective ḍiŋ-ḍiŋ ‘red’ can also be used idiomatically to convey the sense of
‘red-hot’, as in example 28.
146
147
148
149
150
Ibidem.
Ibidem.
Ibidem, p. 318.
Ibidem.
Ibidem.
222
MARK TURIN, BENJAMIN CHUNG
28 dorok-cara-Ø-ta-le
uma-ye
nah-Ø-u-du
run-PSGcome.from.level-sAS-IPP-PCL
wife-ERG
put-sAS-3P-NPT
ḍiŋ-ḍiŋ
baṇi-te
biy-Ø-an.
red
pot-LOC
enter-sAS-3S/PT
He came back [home] running and dove into the red-hot pot his wife
had put ready.151
In the Sindhupālcok dialect of Thangmi, however, a distinction is made between keret
‘red (permanent)’ and jyiŋ-jyiŋ ‘red (temporary)’. In the village of Cokaṭī, the adjective
keret ‘red’ is used for clothes, coloured pens and blood, and jyiŋ-jyiŋ ‘red’ used to describe
the sunset, someone’s face when hot, and irritated or inflamed eyes.
The only other adjective used in Thangmi to convey a sense of colour or hue is
ariŋalya ‘yellow-orange-red’, derived from the Nepali noun ariṇgāl ‘hornet’ on account
of the insect’s golden colouring. Although not widely used, Thangmi speakers assert
that ariŋalya ‘yellow-orange-red’ is a native Thangmi colour word and not a loan. An
example of its use is given in 29.
29 saŋa
peṇey-eŋ-du
sum
oli
măina-ye,
saŋa
ariŋalya
millet
sow-pAS-NPT
three
four
month-ERG
millet red.yellow
tha-Ø-ta-le
min-Ø-du.
be-sAS-IPP-PCL cook-sAS-NPT
Three or four months after it has been sown, the millet turns a goldenyellow colour and is then ripe.152
Similar formulations exist in the more distantly related Magar language. In the
Tanahu dialect, or-cyo ‘a yellow-orange hue’ represents a spectrum of warm colours
while dɦokrot-cʌ is used in the Syangja speaking Magar community for the same range
of pigmentation.153 In Magar, discrepancies of colouration exist between speakers, with
something orange plausibly considered gya-cyo~cʌ ‘red’ and something ‘yellow’ deemed
to be phi-cyo~cʌ ‘green’.154 In Magar, dɦokrot-cʌ and or-cyo are actually verbs that mean
blossoming with a character that is intrinsically ‘charming’ and ‘fresh’.155
30 ort-cyo
suntala
jyap-mʌ156
le.
yellow-ATT orange
tasty-NOM
IMPF
The yellow orange is tasty. (Tanahu dialect)157
Ibidem.
Ibidem., p. 319.
153 Grunow-Harsta, Descriptive Grammar of Two Magar Dialects, p. 275.
154 Ibidem., p. 276.
155 Ibidem., p. 274.
156 Ibidem., p. 255: Grunow-Harsta’s grammar also features [>jyapme] in relation to [jyap-mʌ] and thus appears
to be an alternative realisation.
157 Ibidem.
151
152
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
223
31 gya-cʌ
suntala
jyap-mʌ
le.
red-ATT
orange
tasty-NOM
IMPF
The red orange is tasty. (Syangja dialect)158
In addition to or-cyo/dɦokrot-cʌ ‘yellow-orange hue’, gya-cyo~cʌ ‘red’, and phi-cyo~cʌ
‘green’, Magar contains bo-cyo~cʌ ‘white’ and cik-cyo~cʌ ‘black’. Additionally, nilo
‘blue’ and khailo ‘brown’ are attested in Magar, but these are both direct loans from
Nepali.159
Continuing with Thangmi, the adjectives athaŋ ‘light’ and ukhiŋ ‘dark’ can combine
with postpositions to mean ‘in the daylight’ or ‘in daytime’ and ‘in the dark’ or ‘at night’,
much like in Yakkha, as in examples 32 and 33 below. Younger Thangmi speakers also
use athaŋ ‘light’ as a noun to mean ‘light bulb’, as shown in example 34.160
32 athaŋ-te
caway-eŋ-ta-le,
begale
dese
mi-ye
light-LOC
walk-pAS-IPP-PCL
other
village
person-ERG
niy-eŋ-du
ŋa-to-le,
ṭaye
caway-eŋ-du.
see-pAS-NPT
say-TPP-PCL
night
walk-pAS-NPT
Fearing that if they walked in the daytime they would be seen by people
from other villages, they walked at night.161
33 ukhiŋ-ŋaŋ,
ubo
mi-pali
kiji
icinis-eŋ-du.
dark-inside
white
person-p black
appear-pAS-NPT
In the dark, [even] white people seem to be black.162
34 athaŋ
sat-wa-du-be,
dewa
yo-sa
mi
light
kill-1pg23-NPT-TOP god
look.at-INF
person
kyel-Ø-ta-le
ni-kăi
ci-let-i-n.
come-sAS-IPP-PCL we-PM
CAUS-appear-1pPS-PT
But even though we extinguished the light, a worshipper coming to the
temple saw us and dragged us out.163
Both athaŋ ‘light’ and ukhiŋ ‘dark’ are derived from Indigenous Thangmi verb forms,
athaŋsa ‘to become light’ and ukhiŋsa ‘to become dark’ respectively, examples of which
are given in 35 and 36 below.
158
159
160
161
162
163
Ibidem.
Ibidem.
Turin, Grammar of the Thangmi Language, p. 319.
Ibidem.
Ibidem.
Ibidem.
224
MARK TURIN, BENJAMIN CHUNG
35 athaŋ-Ø-an
ŋa-to-le
priŋ
let-Ø-an,
become.light-sAS-3S/PT say-TPP-PCL outside
appear-sAS-3S/PT
ukhiŋ
nih-Ø-u-no
‘hara
thah-Ø-an?’
dark
see-sAS-3P-3g3/PT
what
be-sAS-3S/PT
ŋa-to-le
duŋ-ŋaŋ
biy-Ø-an.
say-TPP-PCL
within-inside
enter-sAS-3S/PT
Thinking that it was light, he went outside only to see that it was still
dark, ‘what happened?’ he thought to himself, as he went back inside.
36 ṭaye
tha-Ø-ta-le
ukhiŋ-Ø-an,
ni
ariy-i-n.
night be-sAS-IPP-PCL become.dark-sAS-3S/PT we
be.afraid-1pPS-PT
Night fell and it became really dark, and we were frightened.164
By way of contrast, the geographically proximous Dolakha Newar (definitely
endangered, 10,000 speakers, ISO 639-3: new)165 exhibits a larger inventory of colours.
Table 13. Adjectival Verbs in Dolakha Newar166
Post-Nominalisation
Gloss
Stem
wõga-u
wõgar-
green
mwāsa-u
mwāsar-
yellow
twāya-u
twāyar-
white
hẽga-u
hẽgar-
red
phuta-a
phutar-
brown
siya-u
sier-
grey
hāka-u
hākar-
black
In Dolakha Newar, colours exist as “adjectival verbs”167 while simple adjectives exist
as a separate set of words. It is important to note that these words may not correspond
with the Newar language as spoken in Kathmandu, as the dialects of the city and Dolakha
are quite distinct and mostly mutually unintelligible.168
Ibidem, p. 320.
Carol Genetti, A Grammar of Dolakha Newar, Berlin De Gruyter 2007, p. 557; UNESCO Atlas of the World’s
Languages in danger.
166 Genetti, Grammar of Dolakha Newar, p. 195.
167 Ibidem.
168 Ibidem, p. 24.
164
165
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
225
38 āle
hẽga-u
wāsti
ināgu
phi-en
liŋā-i
then red-NR1 clothes
like.this put.on-PART
walk-INF
ma-ji-uju.
NEG-appropriate-3PA
It was not appropriate to put on red clothes and go out like this.169
39 simā
wõga-en
yer-a.
tree
green-PART
come-3sPST
The tree became green (i.e. ‘leafed out’).170
Tibetan (vigorous; 1,172,940 speakers; ISO 639-3: bod),171 one of the most recognizable
languages of the Tibeto-Burman family, is a collection of dialects and speech varieties that
form a linguistic area encompassing Pakistan (Baltistan), Sichuan and Qinghai regions of
China, Sikkim in India, Bhutan, and northern Nepal.172 Within this dialectal variation exist
other Tibetan forms including a literary form as well as slang and “secret languages.”173
The data presented here represent Standard Tibetan and are drawn from Tournadre and
Dorje’s manual and grammar. This data corresponds to what is generally referred to as
the Lhasa dialect and is the variety most commonly spoken in the Tibetan diaspora.174
Table 14. Abridged Colours in Tibetan175
Tibetan
English
kārpo
white
nakpo
black
mārpo
red
sērpo
yellow
cangku
green
ngȫnpo
blue/ green (grass, crops)
gya muk
brown
40 motra nakpo yö’-ngän the tsongpön cī re’
Ibidem, p. 197.
Ibidem, p. 201.
171 UNESCO Atlas of the World’s Languages in Danger.
172 Nicolas Tournadre and S. Dorje, Manual of Standard Tibetan: Language and Civilization: Introduction to
Standad Tibetan (Spoken and Written) Followed by an Appendix on Classical Literary Tibetan, N.Y. 2003, p. 25.
173 Ibidem, p. 26.
174 Ibidem, p. 25.
175 Ibidem, pp. 91,127, 497.
169
170
226
MARK TURIN, BENJAMIN CHUNG
The person who has the black car is the mayor of the district (or alternatively: the owner of the black car is the mayor of the district).176
Syntactically, Tibetan is similar to Nepali in that attributive adjectival phrases are
ambiguous with predicative adjectival phrases unless a demonstrative is used.177 For
example, nyūku nakpo re’ could either signify ‘The pen is black’, or ‘(This) is a black
pen’, but with the use of a demonstrative, the sentence is clarified: nyūku ti nakpo re’
‘This pen is black’ or ti nyūku nakpo re’ ‘This is a black pen’.178 Lexical similarities are
well attested with other Tibeto-Burman languages such as Yolmo, whose colour scheme
(which reportedly only consists of four terms), is cognate with ‘black’, ‘white’, ‘red’,
and ‘blue/green’ in Tibetan.
Most Tibetan colours end with <–po>, a nominaliser, with the exception of cangku
‘green’, even though grammatically speaking, cangku still functions like other colour
terms. Similar to Ladakhi, a close linguistic relative, the term ngönpo covers both ‘blue’
and ‘green’, although the distinction not entirely clear. In Tibetan, while ngönpo is used
to describe both the ‘blue’ sky and ‘green’ plants on earth, generally speaking, ngönpo is
glossed as ‘blue’, especially in the context of Tibetan Buddhism where it is an important
colour in sacred art and material culture.
In Tibetan, colour terms such as ‘purple’ defy easy classification. While marmuk
equates more or less to ‘maroon’ and is used to identify the colour of monks’ robes,
mumen might more accurately equate to English ‘purple’, and literally is the colour of
clotted blood.
Colour mixing is quite productive in Tibetan. To create a dark variety of a colour,
a speaker can combine the colour morpheme of focus without the suffix (i.e. sër- ‘yellow’)
with nāk ‘black’. While we might therefore extrapolate that to acquire a lighter variety
of a colour, a colour morpheme would combine with kārpo ‘white’, this is not in fact
the case, and the term for ‘grey’ is used instead. In Tibetan, colour combining can also
produce additional colours, such as ‘orange’, which would be mar ser ‘red–yellow’. In
sum, Tibetan exhibits a wide palate of colour terms, and the rich Tibetan literary traditions
and religious art of Tibetan Buddhism are suffused with these terms.
Thakali (vulnerable; 6,441 speakers; ISO 639-3: ths)179 is a Tamangic language
with dialects spoken in the Kāli-Gaṇḍaki valley of Lower Mustang district, Nepal.180
Nevertheless, its historic descriptions regarding its genetic affiliations are complicated
and have been contested.181
Ibidem, p. 250.
Ibidem, p. 95.
178 Ibidem.
179 UNESCO Atlas of the World’s Languages in Danger.
180 Stefan Georg, Marphatan Thakali, München 1996, pp. 4, 11; M. Turin, Too Many Stars and Not Enough
Sky: Language and Ethnicity Among the Thakali of Nepal, “Contributions to Nepalese Studies” (24) 2, (1997),
pp. 192–194.
181 Turin, Too Many Stars, p. 192.
176
177
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
227
Table 15. Thakali Adjectives with English Translations (note: originally in German,
our translation)182
Thakali
English
tar2
white
mlaṅ2
black
ur2
yellow
ol1
red
pin2
green/blue
Like many of its linguistic relatives, Thakali adjectives are a distinct class and exhibit
behaviours that are more verbal than nominal.183 In Thakali, colour terms must function
either in a predicative or attributive manner, as they never appear as lone words or
independent descriptors.184
Thakali has four distinct tones, which also account for breathiness and clarity, as well
as intonation. They are denoted by superscript numbers in Stefan Georg’s description of
Marphatan Thakali.185
41 cu2
miná1-e
sipjá1
ol1 mu1.
PDEM
Vogel-GEN
Flügel
rot COP
DEU: Die Flügel dieses Vogels sind rot.
ENG:* The birds’ wings are red.186
*
Our translation from German to English.
42 nakju1 tar2
nu4 –si
mu1.
Hund
weiß
schlafen-CV
COP
DEU: Der weiße Hund schläft.
ENG:* The white dog sleeps.187
*
Our translation from German to English.
43 teptáṭ3
suṅ1
mlaṅ2
la1 –si
p’arki1 –si
Devadatta Mund schwarz machen-CV zurückgehen-CV
DEU: Devadatta zog ein saures Gesicht und ging fort.
ENG:* Devadatta drew a sour face and carried on.188
*
182
183
184
185
186
187
188
Our translation from German to English.
S. Georg, Marphatan Thakali, p. 100.
Ibidem.
Ibidem, p. 101.
Ibidem, pp. 62, 65.
Ibidem, p. 101.
Ibidem.
Ibidem, p. 17.
je4 –ci.
gehen-PRAET
228
MARK TURIN, BENJAMIN CHUNG
44 ṅa1 kon2 ol1 –pa kon2 –la a3 man4 mu1
DEU: Ich trage nicht gerne rote Kleider.
ENG:* I do not like wearing red clothes.189
*
Our translation from German to English.
Examples 41 to 44 are attested while 45 below is ungrammatical and not acceptable
to native speakers. The putative sentence below could be imagined as an answer to the
question, “which dog sleeps? [welcher Hund schläft?],” and while example 45 would be
perfectly acceptable in German,190 it highlights syntactic limitations within Thakali. In
other Himalayan languages, such as like Kulung (see example 2 above) and Thangmi,
colour can be nominalised in this manner.
45 *tar2
nu4 –si
mu1.
weiß
schalfen-CV COP
Der Weiße schläft.
DEU: Der Weiße schläft.
ENG:* The white (one) sleeps.191
*
Our translation from German to English.
Lepcha (definitely endangered, 30,000 speakers, ISO 639-3: lep)192 is spoken in India
(Sikkim and West Bengal), Nepal, and Bhutan.193In common with other Tibeto-Burman
languages, many of its adjectives are derived from verbs. In Lepcha, the prefix <ʔá->
is often attached to the suffix <–m> which modify verbs and render them adjectival.194
However, not all Lepcha colour terms make use of these affixes.
Table 16. Lepcha Colour Adjectives195
Lepcha
English
ʔáhyur
red
ʔánók
black
ʔádúm
white
fungfing
blue
ʔáfong
green
Ibidem, p. 101.
Ibidem.
191 Ibidem.
192 UNESCO Atlas of the World’s Languages in Danger.
193 Heleen Plaisier, A Grammar of Lepcha, Languages of the Greater Himalayan Region (Vol. 5), Brill, Leiden
2006, p. 1.
194 Ibidem, p. 92.
195 Ibidem, pp. 92, 229, 243.
189
190
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
229
In addition, while Lepcha attests a term for ‘light or bright’ ʔákyâng,196 there is
curiously no term for ‘yellow’ in the lexicon. As the Lepcha language possesses ‘blue’,
theoretically Lepcha would be considered to be at Stage V on the Berlin & Kay scale.
However, the absence of ‘yellow’ violates this hypothetical progression.
46 tíngmú-sang
ʔánók
gum.
plainsfolk-PL.H
black
be.AST
Plainsfolk are black, or alternatively, people from the plains are black.197
Yolmo (Ethnologue: vigorous, 10,200 speakers, ISO 639-3: scp)198 is a central Bodish
language comprised of three dialects spoken in central and eastern Nepal.199 Like other
Tibeto-Burman languages, it does not perfectly fit the model proposed by Berlin and
Kay. The data presented here derive from the Lamjung dialect, which has at minimum
an 85% lexical similarity with related dialects.200
Table 17. Yolmo Colour Terms201
Yolmo
English
màrmu
red
nàkpu
black
kárpu
white
ŋómbu
green/blue
Theoretically, Yolmo would fall somewhere between Stage III and V in the Berlin
& Kay model, as there is a discrepancy between ‘blue’ and ‘green’ in the word, ŋómbu.
47 ŋómbu
míi
blue
eye
blue eyes202 (Consultant’s initials: KL; File code: 120304-02)
48 ŋómbu
yìmba
green
COP.EGO
it is green203 (Consultant’s initials: AL; File code: 120209-02 01:07)
Ibidem, p. 92.
Ibidem, p. 57.
198 Ethnologue: Languages of the World.
199 Lauren Gawne, A Sketch Grammar of Lamjung Yolmo, Asia-Pacific Linguistics, National University of Australia,
Canberra 2016, pp. 7–10.
200 Ibidem, p. 9.
201 Ibidem, pp. 47, 62, 78, 91, 136.
202 Ibidem, p. 47.
203 Ibidem, p. 91.
196
197
230
MARK TURIN, BENJAMIN CHUNG
As evidenced in examples 47 and 48, Yolmo ‘blue’ and ‘green’ are covered by the same
lexeme. In addition, there is no documented term for ‘yellow’ in Yolmo. As previously
noted, ‘blue’ and ‘green’ hues can be perceived simultaneously, which may account for the
interchangeability of this term.204 Nonetheless, the absence of distinct ‘blue’ and ‘green’
complicates the language’s placement on the Berlin and Kay progression that dictates
that for ‘blue’ to appear in a lexicon, ‘green’ and ‘yellow’ must both be present prior.
Thus, Yolmo is again another language that helps form this loose trend of Tibeto-Burman
languages that have a ‘blue’ and ‘green’ discrepancy. Furthermore, Yolmo shares cognates
for ‘black’ and ‘white’ with its sister languages, as seen in examples 49 and 50 below.
49 mì
nàkpu=la
pèmpiʑa
tɕí=ki
tɕà
person black=DAT woman
one=ERG
tea
To the black man a woman carried and gave tea.205
(Consultant’s initials: AL; File code: 101006-01)
kyòŋ-ti
carry-PERF
tér-sin
give-PST
50 sá=la
bàltiŋ
kárpu
tɕéemi
tɕíi
dù
ground=LOC
bucket
white
small
one
COP.PE
206
A small white bucket is on the ground.
(Consultant’s initials: AL; File code: 101010-01 11:15)
Outliers in the Tibeto-Burman Family
While the above section addresses Tibeto-Burman languages that are (for the most
part) relatively regular in how they express colour, the following section is devoted to
outliers in the Tibeto-Burman family with regard to the expression of colour terminology.
These languages address colour in ways that are quite unlike their linguistic relatives and
exhibit striking conceptual and lexical differences to their counterparts. In these languages,
the approach to colour is more complex, and perhaps unsurprisingly, these languages also
profoundly challenge Berlin & Kay’s theory of universal colour progression.
Gyarong
Gyarong, also known as rGyal rong in Written Tibetan, has a total of eleven colour
elements.207 As a consequence of its geographical location in Sichuan, China, Gyarong
exhibits multiple loans from both Chinese and Tibetan. However, similar to other TibetoHurvich and Jameson, Theory of the Light Sense, 1964; as cited in Kay and McDaniel, Linguistic Significance
of the Meanings of Basic Color Terms, p. 620.
205 Gawne, Sketch Grammar of Lamjung Yolmo, p. 125.
206 Ibidem, p. 78.
207 Nagano, Gyarong Color Terms, pp. 99, 101.
204
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
231
Burman languages surveyed in this article, many Gyarong colours function as verbs.
Jacques also highlights that Gyarong colours (as well as other descriptive concepts) are
ideophonic and thus relate to physical and cognitive perception.208 While a classic example
of auditory ideophones would be onomatopoeia, visual cues and triggers also constitute
this sub-field of sensory semantics. Another unusual and grammatically interesting feature
of Gyarong is that its speakers can combine colours from languages that have provided
loans (namely Chinese and Tibetan) to create entirely new lexemes in their own language,
Gyarong.
Colours in Gyarong can be thought to represent a spectrum in which each increment has
set boundaries. There is evidence that these gradients correlate with the place of articulation
of consonants, with lighter colours correlating more with alveolars and retroflexes while
darker colours are associated more with velars and uvular fricatives.209 Gyarong appears
to be a Stage II language with the exception of ‘grey’. Nagano, a Japanese scholar of
Gyarong, also references the Berlin & Kay proposal that ‘grey’ may function as an outlier
with unpredictable focus.210
‘Grey’, which is either kə pki or kə phyi in Gyarong, falls between the spectrum of
kə pram ‘white’ and kə nak ‘black’,211 with kə- functioning as a prefix that appears only
before certain colours. While kə pram, ke nak, and kə phyi may all have roots in ProtoTibeto-Burman, kə pki has no known extant cognates.212 These Gyarong terms behave
as verbs, which cannot be said of all Gyarong colour terms, and – through the Munsell
colour system code for colours (hereafter MC) – they are represented as N9–N6 and
N1.5.213 Additionally, in the Japhug variant of Gyarong, sɯŋ and zɯŋ both mean ‘white’,
and both terms can be used to describe an old person’s hair colour.214 Continuing along
this light-coloured spectrum, ʂɯŋ and x̣ɯŋ from the same Japhug variant of Gyarong can
both mean ‘clear’, in the sense of the sky being ‘clear’. The first term, ʂɯŋ, has a more
metaphoric meaning and can also describe an action, while x̣ɯŋ is a more literal term,
which would be used to describe ‘the colour of dead skin’.215
In common with all Gyarong colours, ‘red’ covers a gradient. The most accurate term
is kə wu rne, which includes 10RP, 4R (which is simply ‘red’), 7R and 6RP in MC.216
We would like to thank one of our anonymous reviewers for directing us to this additional dialect of Gyarong
to include in this study: G. Jacques, Ideophones in Japhung (Rgyalrong), “Anthropological Linguistics,” 55 (3)
2013, p. 263.
209 Jacques, Ideophones, p. 267.
210 Nagano, Gyarong Color Terms, p. 106; as cited in Kay and McDaniel, Linguistic Significance of the Meanings
of Basic Color Terms, p. 620.
211 Nagano, Gyarong Color Terms, p. 100.
212 Ibidem, pp. 101–102.
213 Ibidem, p. 100.
214 Jacques, Ideophones, p. 267.
215 Ibidem, p. 267.
216 Nagano, Gyarong Colour Terms, p. 100.
208
232
MARK TURIN, BENJAMIN CHUNG
It appears that kə wu rne is a pure Gyarong word with clear Proto-Tibeto-Burman origins,
and that like ‘black’, ‘white’, and ‘grey’, it acts like a verb.217
By contrast, the Gyarong term li ṭhi represents a more ‘orangey red’, identified as
10R, 4YR, 8YR, and 2Y in MC,218 although this word is a loan from Written Tibetan,
li khri, which means a ‘minium, red lead’.219 Jacques observes that in the Japhug variety
of Gyarong, χɑŋ also means ‘slightly orange’ like ‘the sky at daybreak’.220
Although Berlin & Kay’s theory stipulates that the linear progression of colour terms
leads directly to ‘red’ after the ‘white’ and ‘black’, it does not specify how many ‘reds’
can exist in any given language; and as we know, multiple varieties of colours appear in
other Tibeto-Burman languages as well (i.e. Thangmi with multiple ‘reds’). Gyarong is
interesting in that it possesses two terms for the same gradient, a peculiarity that conforms
to the extension of the theory with fuzzy sets. In Gyarong, ‘red’ itself is encompassed
by kə wu rne, and would likely be classified as a theoretical ‘red’.221 Given that li ṭhi
is a loan, we may assume that its addition to the lexicon of Gyarong is secondary. For
this reason, the existence of two terms for the same gradient raises the question of what
the borrowed term li ṭhi encompasses that the native term kə wu rne does not.
At this point, at least according to a canonical reading of Berlin & Kay’s theory,
either ‘yellow’ or ‘green’, but not both, may appear in a lexicon to qualify as a Stage III
language. Both ‘yellow’ and ‘green’ are attested in Gyarong, with a larger spectrum for
‘green’ than for ‘yellow’, although both terms are loaned from neighbouring languages.
In Gyarong, ‘yellow’ is sii po, directly borrowed from Written Tibetan ser po222 and
manifests as 5Y and 8Y in the Munsell code format (MC). The first ‘green’ interval
is ǰaṅ ku, which is represented by 3GY, 8GY, 3G, 9G, and 5BG in MC,223 and which
stretches into ‘visual yellow’ and ‘blue’ territories.224 The term ǰaṅ ku also derives from
Written Tibetan, originally ljiang khu, which is in turn derived from the term for “(pine)
sprout.”225
‘Green’ and ‘blue’ fade together in Gyarong, and the next section of this analysis
focuses on ‘blue’ and its variants. The first term for ‘blue’ is arguably the most fascinating
colour term in the Gyarong language – laṅ kar 226 – and represented by 10BG, 5B, and
10B in MC, which the PCCS lists as “blue green, greenish blue [and] blue” respectively.227
Aside from its position in the ‘green-blue’ range, the etymology of laṅ kar is unusual.
The first syllable is derived from Chinese lan, meaning ‘indigo’ while its second syllable
217
218
219
220
221
222
223
224
225
226
227
Ibidem, p. 101.
Ibidem, p. 100.
Ibidem, p. 101.
Jacques, Ideophones, p. 267.
Kay and McDaniel, Linguistic Significance of the Meanings of Basic Color Terms, p. 622.
Nagano, Gyarong Color Terms, p. 101.
Ibidem, p. 100.
Ibidem.
Ibidem, p. 101.
Ibidem.
Ibidem.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
233
derives from Written Tibetan dkar ‘white’.228 Somehow, two terms from different languages
synthesized into a new term in the lexicon of a third language. Nagano notes that, “this
particular ‘blue’ is expressed as ‘whitish indigo’.”229 How such a process occurred, and with
what motivation, lie beyond our current analysis and firmly in the realm of conjecture. To
add further complication, another term exists for ‘whitish indigo’, ṅon kya.230 Its etymology
is purely Tibetan, combining the word sngon po ‘blue’ and skya ‘gray, faint’.231 No data
is available to indicate which of the terms is used in which contexts, and whether they
exist in free variation or rather in some form of complementary distribution.
An additional Gyarong word for ‘blue’ is ṅon po, derived directly from Tibetan, with
an MC correspondence of 3PB.232 When compared to ṅon kya, ṅon po may exhibit the
highest degree of membership to ‘blue’ as it is less varied on perceivable and quantifiable
colour, although it may be that speakers of Gyarong view and categorize colour in
a manner that does not necessitate the presence of a ‘true blue’ per se.
The terms ser muk ‘brownish gold’ and laṅ ‘indigo’ are also attested in the Gyarong
lexicon.233 While laṅ is a loan from Chinese as noted above, ser muk also may indicate
a cultural perspective in colour perception. While the PCCS description for ser muk is
technically ‘charcoal’ with an MC of 7.5YR 5/8, it is described as ‘brownish gold’ by
Nagano.234 The origins of this synthesized colour term come once again from Written
Tibetan, with the element ser deriving from gser ‘gold’ and muk from smug ‘dark bay,
cherry-brown, brownish’.235
Chepang
The Chepang language spoken in Nepal (vulnerable, 36,807 speakers, ISO 639-3:
byh, cdm)236 expresses colour very differently to other Tibeto-Burman languages in the
region. While colours do exist in Chepang, they are usually implicitly connected to what
they describe, or they reflect the method by which the described object obtained its
colouration. For example, dut is used to describe only ‘white fluids’ while its counterpart
phir is used only for ‘things, especially white clothing’.237 Interestingly, dut is a Nepali
borrowing.238
Ibidem.
Ibidem.
230 Ibidem.
231 Ibidem.
232 Ibidem.
233 Ibidem, pp. 100–101.
234 Ibidem, p. 100.
235 Ibidem, p. 101.
236 UNESCO Atlas of the World’s Languages in danger.
237 Ross Ch. Caughley, Dictionary of Chepang: A Tibeto-Burman Language of Nepal, Pacific Linguistics, National
University of Australia, 2000, p. 486.
238 We thank one of our anonymous reviewers for advising us for the origin of this term.
228
229
234
MARK TURIN, BENJAMIN CHUNG
The elements gal- or gaw- are most commonly used to describe things that are
‘black’. However, they take different forms depending on the nature of what is described.
Table 18. ‘Black’ in Chepang239
English
Chepang
clouds
gal.tiŋ.ki.taŋ.kə
pot
gal.taŋ.kwar
very
cik.cak, tik.tak, hak.ca.də
coating of skin
norʔ
with anger
khen gal-
In Chepang, a distinction is made between that which is ‘black’ and that which is
‘blackened’. In only one instance, gal is attested, while in the other examples, contrasting
morphemes are used.
Table 19. ‘Blackened’ in Chepang240
English
Chepang
bruised
hno-
corn tassels
gal.koy.rəʔ, gal.gən.də.rəŋ
grain, mat, etc.
puŋ-
tubers
ŋəlʔ-
It is worth noting that the majority of the terms to describe things that are ‘blackened’
relate to agriculture and forest products, such as tubers, grain and corn. In addition to
these descriptors, Chepang attests a distinct word for ‘blackish’, which manifests as either
<jhik-> or <ŋələʔ->241 and the term for ‘black’ relating to ‘grime, from burning oil’ is
nalh, best translated with the English noun ‘blackness’.
Like ‘black’, the colour ‘white’ has many variations in Chepang with most having
some biological significance:
239
240
241
Ibidem, p. 319.
Ibidem.
Ibidem.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
235
Table 20. ‘White’ in Chepang242
English
Chepang
coating on newborn baby
tuk.rəyʔ
cumulus cloud
romʔ.mus
earth used for whitewashing
kə.mi.ro (Nep.?)
fluid
dut (Nep.?)
thing
phir
The Chepang verb ‘to be white’ has its own paradigm with the stems pham- or
bham- and three possible shapes or positions: coiled, stretched out, or well-shaped.243
In addition, while reduplication is apparent in this paradigm, it is only attested in
one form:
Table 21. ‘To be White’ in Chepang244
English
Chepang
something that is coiled up
bham.kwayʔ.kwayʔ
something that is spread out
bham.blaŋ
something that is well-shaped
bham.jheŋ.teŋ
One final term relates to ‘whiteness’ in Chepang: norʔ.kliʔ. This word can best be
translated as ‘whitish faecal discharge’, once again highlighting the biological aspect to
Chepang colour terminology.245
For ‘red’, Chepang has a rather small repository of terms, most of which relate to
the sun or to fire. The main stem is du-246 and reduplication resurfaces in the term for
‘sky, sunset’.
242 Ibidem, p. 486. We thank one of our anonymous reviewers for advising us of the possible Nepali origin of
these terms for ‘white’.
243 Ibidem.
244 Ibidem
245 Ibidem.
246 Ibidem., p. 426.
236
MARK TURIN, BENJAMIN CHUNG
Table 22. ‘Red’ in Chepang247
English
Chepang
flames
dili bili
sky, sunset
du.he.re.re
appearing (via the sun)
du.taŋ.kwar
sap
wəyʔ
sky or clouds
du.syo.paŋ
The one variation for ‘reddening’ in Chepang also relates to the sun, in which the
stem manifests unmodified as kwal.hal.ya.248
Table 23. ‘Reddening’ in Chepang249
English
reddening (of the sun)
Chepang
he.ray.lə
In Chepang, the colour term for ‘green’ exhibits one stem with no alternation: <pli->.250
Unsurprisingly, and continuing in the biological vein, most words that derive from <pli->
relate to water or foliage, as these domains are where the colour is most likely to occur
in the natural world.
Table 24. ‘Green’ in Chepang251
English
247
248
249
250
251
Chepang
deep
pliwi
deep, dark
pli.ma.rit
deep (water)
kwiŋ, wis
deep (relating to growth)
pliti.nik
timber etc.
syoyʔ
tree
jiŋʔ-
very
pli.jhəyŋ, pli.layŋ
very (relating to trees)
jhuyŋʔ
Ibidem.
Ibidem.
Ibidem.
Ibidem.
Ibidem, p. 374.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
237
In Chepang, the concept of ‘yellow’ has three distinct stems: the colour itself,
something that has acquired the colour, and the process in which something acquires
the colour. All but two forms relate in some manner to tree leaves.
Table 25. ‘Yellow’ in Chepang252
English
Chepang
‘(become – leaves)’
ʔəmhə(ʔ)-
leaves
ʔepeʔ-, ʔemheʔ-
very
yar.ba.li
Table 26. ‘Yellowed’ in Chepang253
English
Chepang
‘(become – clothes)’
klip-
yam leaves
ʔemheʔ
Table 27. ‘Yellowing’ in Chepang254
English
leaves with dryness
Chepang
buh.rayʔ
In Chepang, the colour ‘blue’ is the same as the colour ‘green’: <pli->.255 Without
additional information, this co-occurrence violates the most traditional and narrow
reading of Berlin & Kay’s theory of colour, as Chepang exhibits discrete words for
‘grey’ and ‘purplish’ (which are addressed below). However, as Kay & McDaniel’s
revised theory proposes, the emphasis on loose categories or ‘fuzzy sets’ enables the
connection between ‘blue’ and ‘green’ to be viable in Chepang and not contradict the broad
strokes of the theory. In addition, Hering’s understanding of the simultaneity of ‘blue’
and ‘green’ perception may also indicate a deeper connection between these colours in
this language.
Although ‘brown’ does exist as a colour in Chepang, it is determined as either ‘light’
or ‘dark’ in nature, and never as a neutral variant.
252
253
254
255
Ibidem, p. 490.
Ibidem.
Ibidem.
Ibidem, pp. 320, 374.
238
MARK TURIN, BENJAMIN CHUNG
Table 28. ‘Brown’ in Chepang256
English
Chepang
(light) brown
ŋalʔ-
(dark) brown
phut-
The last colours that we survey for Chepang are ‘purple’ and ‘grey’, as neither ‘pink’
nor ‘orange’ exist in the language. Even the status of ‘purple’ is questionable, as although
there is a word to describe something ‘purplish,’ there is no discrete word for ‘purple’
itself. As for ‘grey’, variations of the word are attested much like the other colours, with
specific lexicon devoted to the process of ‘greying’.257
Table 29. ‘Purple’ in Chepang258
English
‘purplish’
Chepang
ŋalʔ-
Interestingly, the word for ‘purplish’ also means ‘light brown’, although the exact
nature of this semantic relationship is unknown.
The main stem for the colour ‘grey’ is <brus->, and most references relate to hair.259
Table 30. ‘Grey’ in Chepang260
English
Chepang
hair
brok-
and white (of clouds)
phut hoyo.bhoyoʔ
dusty
phut-
or white streaked hair
myang.brok
Ladakhi
Despite deep historic and cultural influences from Tibet itself, the Ladakhi language
is quite distinct with five dialects that vary in phonology, grammar, and lexicon. The
majority of Ladakhi speakers live in Jammu and Kashmir.261
256
257
258
259
260
261
Ibidem, p. 422.
Ibidem, pp. 422, 374.
Ibidem, p. 422.
Ibidem, p. 374.
Ibidem.
Sanyukta Koshal, Ladakhi grammar, Delhi 1979, p. 3.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
239
Dollfus offers an exemplary and fascinatingly intricate portrait of the complex
relationship Ladakhis have with their colour system and asserts, “For Ladakhis, a colour
does not exist per se, but only when associated with the animate or inanimate object this
colour describes. A colour exists within a context, not on a chart.”262 Dollfus suggests
that such a perspective may be a result of the historic religiosity of the region and of
Ladakhis in particular in regard to Tibetan Buddhism, which uses colour thematically
and metaphorically in tantric practices and to structure its worldview.263
Dollfus also demonstrates that the concept of ‘colour’ itself is highly variable in
Ladakhi, with six lexemes in existence to cover the very idea of ‘colour’.264
Table 31. Colour in Ladakhi265
Ladakhi
English
Examples
mdog/ dog
colour; complexion; appearance;
look
tshon
dye; pigment; paint
tschon chen ‘a colour that never fades’;
msthon bkra/ tshon khra ‘coloured/colour’
(i.e. movies or photographs)
tshos
dye
tshos ‘tshos cas/ tshos gtang cas ‘to dye’;
tschos (sogs) dog ‘a (fully) ripe colour’ in
reference to fruit;* today this term is often
replaced with rang
rang (Hin.)
colour; paint; dye; pigment
could pertain to hair dye or dye in dried
(particularly artificial substances) fruit
kha
Compounded to create other
colour terms (i.e. glo kha ‘lung
colour’)
*
**
mchin kha ‘liver colour’; g.yu kha ‘turquoise
colour’; In Tibetan, it can also include
the qualities of lustre and sheen; kha dog
‘colour, attitude, or viewpoint/ perspective’;
at one point, it may have only been used for
the colour of some wet surfaces**
Pascale Dollfus, Using Colours in Ladakh, p. 277.
Y. Nagano, An Analysis of Tibetan Colour Terminology, “Tibeto-Burman Studies” (1), Tokyo 1979.
Additionally, dogor kha can combine with a seemingly endless number of nouns to
create a unique colour term (i.e. ro dog ‘corpse colour’),266 and with additional morphemes
262 Pascale Dollfus, Perceiving, Naming, and Using Colours in Ladakh, “The Tibet Journal” (40) 2, (2015),
p. 262.
263 Ibidem, p. 274.
264 Ibidem, p. 263.
265 Dollfus, Using Colours in Ladakh, p. 263.
266 Ibidem, p. 265.
240
MARK TURIN, BENJAMIN CHUNG
that may alter colours further. It is interesting to note that Ladakhis tend to privilege
‘glossiness’ over ‘shade’ as a result of cultural opinions of beauty and fortune.267 While
mdangs can ‘having brightness’ equates to possessing beauty and ‘shining’, someone
who is described as mdangs med ‘without brightness’ would be considered buffoonish
and dull.268 Such terms of brightness are also used to describe horses.
Table 32. Shades in Ladahki269
Ladahki
English
sprin pa
cloud
du ba
smoke
rdul
dust
khug sna
mist
nyi ma
sun
grib pa
shadow
snag ba
lustre
mun pa
darkness
Table 33. Shapes and Additional Descriptors in Ladakhi270
Ladakhi
267
268
269
270
English
ring
long
thung
short
mtho
high
dma’
low
lham
square
zlum
round
phya le ba
level (even)
phya le ba ma yin pa
not level (uneven)
Ibidem., pp. 271–273.
Ibidem., p. 273.
Ibidem, p. 263.
Ibidem.
241
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
Table 34. Additional Colour Modifiers in Ladakhi271
Ladakhi
English
skya/skya bo
pale, light, faint (i.e. ser skya ‘pale yellow’; sgno skya ‘a light blue’)
nag/nag po
a dark colour (i.e. sngo nag ‘blue black/ dark black’; ser nag ‘a dark
yellow’)
rkyang/skyang
homogeneity/totality (i.e. sngon po sngo rkyang ‘completely and totally
blue’)
khra/ khra bo
technically a combination of black and white (see below)
Within this collection of Ladakhi colour modifiers, some are of particular interest.
For example, although black and white are polarized, Dollfus attests that they are not
“achromatic”.272 In practice, this means that nag skya is ungrammatical and does not signify
“a very intense black” while in fact dkar skya is considered ‘off-white’.273 Additionally,
khra/khra bo has a complex and subtle meaning. Although its base form is a combination
of the terms for ‘black’ and ‘white’, when used in combination with other colours, khra/
khra bo denotes different properties pertaining to the colour that it modifies, i.e. dmar
khra ‘red-spotted’ and dkar khra ‘piebald but predominantly white’.274
Despite the variety of modifiers and colours attested in Ladakhi, Dollfus suggests that
Ladakhi speakers always insist their language has only four colours: dkar po ‘white’, ser
po ‘yellow’, dmar po ‘red; also expressing saturation’, and sngon po ‘blue’.275
51 pu-mo-gun-ni
ṭho-e
Noun-Pl.-Gen.
P.P.-Gen.
girls
among
The face of the elder sister
52 pu-mo-romN.-Qul.Adj.
girl
The tall fat
ə-če-yi
rdoη-kər-po
Noun-Gen.
N.-Qul.Adj-Dir.
elder sister
face-white
among the girls is white.276
po-kər-porηi-mo
-Qul.Adj.Qul.Adj.
fat
white
white girls …277
duk
be Pr.V.to
-gun ...
-Pl.Suf.Dir …
tall
In Ladakhi, these four colours are considered pure and unmixed. In addition, two
other colour terms are attested and are generally unspecified: nag po ‘dirty, black’ and
271
272
273
274
275
276
277
Ibidem,
Ibidem.
Ibidem.
Ibidem.
Ibidem,
Koshal,
Ibidem,
p. 265.
pp. 264–265.
Ladakhi Grammar, p. 88.
p. 145.
242
MARK TURIN, BENJAMIN CHUNG
smug po (which is attested as ‘the colour of clotted blood’). The Ladakhi term smug
po encompasses a range of colours and tints that include ‘dark maroon, dark brown,
dark bay or any dark grey colour, reddish brown, cherry-brown, maroon, brown, purple,
purplish, etc.’ and although it is never used to describe a bold colour, it is used to express
deepness.278 In Ladakhi, smug thig, a counterpart to smug po, is considered the colour
of bruises (thig translates as to ‘drop’).279 The prefix smug also appears in Gyarong,
meaning ‘dark’, which may highlight a relationship to more conservative or archaic
varieties of Tibetan.280
53 ləl-ki-lə
Noun-Dir.
Red Fort
Red Fort
ək-bər-rgyəl-po-e
Noun Noun Gen.
Proper Name King
is built by King Akbar.281
rt͜siks-khən
V.-St.-Deri-Suf.
built
yin-nok.
V. to be-PRES
The term ləl appears in example 53 above, but dmar po is attested as ‘red’ in Ladakhi.
It is highly likely that ləl is a borrowing from Hindi or Nepali, both of which languages
have the word lāl for ‘red’.
Additionally, in his grammar, Koshal notes that ‘green’ in Ladakhi is lǰəη-khu, which
is cognate with Tibetan cangku and Gyarong ǰaṅ ku.282 Dollfus specifies that ljang ku
(her transliteration) is mostly used for description in religious contexts and for artificial
greens.283 However, the nominaliser <–pa> can also appear as a suffix on ljang, meaning
“small green plant or blade of grass,” whereas the pan-Tibetan nominaliser <-po>
cannot, all the more intriguing given that ljang ku is never used to describe natural
greenness.284
Koshal observes that abstract adjective-like nouns such as colours can undergo
suffixation with <-čhə> to slightly change their meanings and solidify themselves as
nouns.285 For example, kər-po ‘white’ and nək-po (as transcribed by Koshal) become
‘whiteness’ kər-čhə and nək-čhə ‘blackness’, respectively. Nevertheless, the distinction
between adjectives and nouns in Ladakhi is not clear, and abstract noun suffixation may
be viable with both noun and adjective roots.286
In addition, the suffix <-čo> can make adjectives (which includes nouns as these
classes are blurred) into verbs.287
278
279
280
281
282
283
284
285
286
287
Dollfus, Using Colours in Ladakh, p. 264.
Ibidem.
Nagano, Gyarong Color Terms, p. 101.
Koshal, Ladakhi Grammar, p. 189.
Nagano, Gyarong Color Terms, p. 100; Tournadre and Dorje, Manual of Standard Tibetan, p. 127.
Dollfus, Using Colours in Ladakh, p. 270.
Ibidem.
Koshal, Ladakhi Grammar, p. 56.
Ibidem.
Ibidem., p. 181.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
243
54 nək-po-čoblack do
(to) Blacken (i.e. to make black)288
Colour terminology is further expanded into natural and biological worlds, finding
connection with both Ladakh’s surrounding flora, fauna (including livestock), the earth,
and Lakadhis’ health and wellness.
Table 35. Colour in Plant Names289
Ladakhi
English
sro lo dmar po
red sholo (red flowering succulent) Rhodiola himalensis
sro lo ser po
yellow sholo (pale yellow flowering succulent) Rhodiola imbricata
bong nga nag po
black aconite Aconitum violaceum
bing nga dkar po
white aconite Aconitum heterophyllum
Table 36. Colour in Animal Names290
Ladakhi
English
smug stag
lit. purple-brown tiger (i.e. clouded leopard) Neofelis nebulosi
thang dkar rgod po
lit. white breasted [and] wild (i.e. Egyptian vulture) Neophron
percnopterus
khrung khrung ske nag
black-necked crane Grus nigricollis
bya khra bo
lit. piebald bird (i.e. magpie)
Table 37. Colour Terminology Pertaining to Livestock291
Ladakhi
English
be lu
a white goat with two dark stripes on its head
rag pa
a goat with yellow hair around its eyes
ngang pa
a goat with fawn hair on its head and on its backbone
mgo nag
a goat with a black head
mgo smug
a goat with a brown head
ser mgo
a goat with a yellow head
288
289
290
291
Ibidem.
Dollfus, Using Colours in Ladakh, pp. 264, 273.
Ibidem, p. 264.
Ibidem, pp. 267–268.
244
MARK TURIN, BENJAMIN CHUNG
Table 37. (cont.)
Ladakhi
English
she lo
a goat with a spotted head “like a meadow full of blossom”
khra bo
‘piebald’
kham pha
‘brown or beige’
rog po
black (yak hair or raven plumage)
nag po
black (sheep wool)
sngon po
grey for a horse/ “a watered-down version of true black”*
*
Dollfus, Using Colours in Ladakh, p. 271.
Table 38. Colour in Minerals292
Ladakhi
English
lcags rdo dmar po
red ironstone (red hematite)
brag ri smug po
basalt
Table 39. Artistic Paints & Pigments in Ladakhi (Plants & Minerals)293
Ladakhi
English
mtschal rgod
cinnabar (derived from wild vermillion)
rgya mtshal/ li khri
‘red lead’; an artificial variety of orange-red from vermillion
(imported from China and India)
dong ros/ btso ma
‘orange-yellow realgar’ from arsenic sulphides
ba bla
a yellow variant from arsenic sulphides
ka rag
white from calcium
skag
Lac dye from the fluid left on twigs by small reddish insects
Laccifer laca and which creates an equally reddish colour
rams (counterpart to skag)
indigo Indigofera tinctoria
We may note that in Gyarong, li khri is used identically and corresponds to the
same colour in Ladakhi.294 Such cognate connections are not surprising given historic
relationships and the widespread diffusion of Tibetan culture in Ladakh and across
the surrounding regions.
292
293
294
Ibidem, p. 264.
Ibidem, pp. 265–266.
Nagano, Gyarong Color Terms, p. 101.
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
245
Table 40. Festivals with Colour Terminology in Ladakhi295
Ladakhi
English
‘Green Gods Festival’ that occurs in early summer
sngo lha
Table 41. Colour in Diseases296
Ladakhi
English
mig ser
lit. yellow eyes (jaundice)
mdze dmar
red leprosy
Table 42. Bodily Terms Combined with Colour Terms297
Ladakhi
English
ser po sngon po chas cas
‘to go yellow and blue’/ to look ill
rgyu dkar
φ white gut
rgyu nag
φ black gut
ya ma dkar po
φ white sinuses
ya ma nap po
φ black sinuses
rtsa dkar po
‘white channels’/ nerves and tendons
rtsa dmar po
‘red channels’/ veins and arteries
khu bar dkar po/ khams dkar po
‘white fluid or element’/ sperm
khu bar dmar po/ khams dmar po
‘red fluid or element’/ menstrual blood
φ – indicates a culturally-relevant medical distinction.
sηon-po
duk.
55 lčəη-mə
N.-Sing-Dir
Qul.Adj
V. to be-PRES
tree
green
The tree is green (on the basis of seeing it).298
We may note that in this context, sηon-po ‘green’ is cognate with Gyarong sngon po
‘blue’ and with Tibetan ngȫnpo ‘blue/green’.299 In the Tibetan context, ngȫnpo is used for
vegetation, which coincides with its usage in Ladakhi where it can also encompass more
295
296
297
298
299
Dollfus, Using Colours in Ladakh, p. 271.
Ibidem, p. 264.
Ibidem, pp. 273–274, 277.
Koshal, Ladakhi Grammar, p. 186.
Nagano, Gyarong Color Terms, p. 101; Tournadre and Dorje, Manual of Standard Tibetan, p. 497.
246
MARK TURIN, BENJAMIN CHUNG
colours than just ‘green’ and ‘blue’ including reds, purples, and greys.300 In Gyarong,
the meaning ‘blue’ is seemingly more prevalent.
56 khoη-ηi
III.P.-Sg.-Gen
his
His horse is
ṣtə
nək-po
yot.
N.Dir
Qul.Adj.
V. to be-PRES
horse
black
black (based on direct knowledge).301
Traditional Ladakhi art practices and Buddhist customs offer rich domains of culturallyspecific colour usage and resonance. While the semiotics of these colours remains a point
of historical debate among artists and theologians, Dollfus argues that colour is rooted
in Tibetan astrology with a supporting literature that dates back to the 15th century.302
Given such a deep tradition, Ladakhi artists have created their own jargon and expanded
the dimensions in which colour can be perceived, particularly in painting. By mixing
and grinding dyes and pigments, Dollfus suggests that Ladakhi painters have created
additional levels of colour specification that are unique to their craft. These colours can
be subdivided further into base colours and finishing colours that are painted on after
an initial coat.303
In the case of what in English would be referred to as ‘blue’ and ‘green’, a base
coat, mthing corresponds to English ‘azurite blue’, while spang or mdo spang corresponds
to English ‘malachite green’. Additional varieties of these colours exist in Ladakhi, all
gradient of shades which are determined by how finely the minerals are ground:
Azurite gradient: [lightest] sngo si >> sngo sang >>mthing shul >> mthing ‘bru [darkest]
Malachite gradient: [lightest] spang si >> spang skya >> spang >> spang smug [darkest]304
Interestingly, while Ladakhi ‘bru means ‘grain’, it is used to represent the darkest
shade of azurite blue in such constructions.305
As outlined above and painstakingly documented by Dollfus, all of these colours
also have a spiritual significance and may be read as a meta-social analysis of gender
and class dynamics. From the 18th-19th century onward, Ladakhi cultural texts reference
eight main colours: seven that represent male/man (mthing, ljang ‘green’, mtschal, li,
skag, ba bla, and rams) and one that represents female/woman (dkar or ka rag). From
these, other colours – sons or bu – are produced.306 The combination of the eight prime
colours creates such ‘sons’, including dmar skya ‘pale red’ and mi sha ‘human flesh’,
Dollfus, Using Colours in Ladakh, pp. 270–271.
Koshal, Ladakhi Grammar, p. 186.
302 Dollfus, Using Colours in Ladakh, p. 265.
303 Ibidem, p. 266.
304 Ibidem, p. 265.
305 David Jackson and Janice Jackson, Tibetan Thangka Painting: Methods and Materials, Boulder, CO (USA)
1984, 91 et seq.; as cited in Dollfus, Using Colours in Ladakh, p. 265.
306 Dollfus, Using Colours in Ladakh, p. 266.
300
301
COLOUR TERMS IN TIBETO-BURMAN LANGUAGES
247
offspring specifically of mtshal gyi bu ‘vermillion’ as well as mchin kha ‘liver colour’,
glo kha ‘lung colour’, ljang ser ‘yellow green’, and rams se ‘indigo’. Two additional
sisters ‘sring mo’ are worth mentioning, ja kha ‘tea colour’ and dud kha ‘smoke’.
Colour terminology in Ladakhi is deeply cultural and cannot be fully understood
without an understanding of the socio-religious context in which the community live. As
a case in point, Buddhist monks are referred to as ser po, or ‘the yellow ones’ because
of their attire, in contrast to common folk who are referred to as skya or ‘grey ones’,
even though their ropes are conventionally more ‘red’ than ‘yellow’. In Ladakhi, Buddhist
monks are also referred to as ngur smrig ‘dzin pa/ nur smrig ‘chang ba, which translates
as ‘reddish-yellow/ saffron robe wearers’ despite not making use of the Ladakhi term for
‘yellow’ or ‘saffron’, kur kum, in the description.307 Dollfus suggests that the use of the
colour term ser po comes from the symbolic significance of the colour ‘yellow’, which
is synonymous with gold, merit, wisdom, and yak butter – the favorite and most useful
butter – thus emphasizing the status of monks.308 Yet, ‘yellow’ is also used to refer to
outsiders and can indicate a negative attribute: mig ser po ‘yellow eyes’ and mgo po
‘yellow head’ can refer to Westerners of any hair or eye colour, while ser sna ‘yellow
nose’ can convey the sense of jealous or greedy.309
Despite the extraordinary detail of systematic adjectival description attested in the
Ladakhi language, Dollfus suggests that Ladakhi society is not as overtly based around
colour as one might expect from such terminological effervescence.310 Ladakhi expressions
that use colour terms are curiously limited; and in everyday usage, colour is not a primary
descriptor, and remains firmly secondary to size or shape. Having noted this, colour is
nonetheless prevalent in Ladakhi literature and is frequently deployed in metaphoric
constructions, reiterating the culturally salient role of colour in Ladakhi language and
society.311
Conclusion
The complexity and diversity of the Tibeto-Burman language family offers us a pathway
to better understand the anthropological, linguistic, and cognitive commonalities between
related cultures and peoples. Colour is one such pathway, showing both unexpected
linkages and pronounced differences across and between Himalayan languages. In the
scope of our survey, linguistic patterns arise that appear to violate aspects of Berlin &
Kay’s initial and modified colour paradigms. These results are significant as they contribute
a new perspective to a process often thought to be universal. These commonalities also
provide further insight to the characteristics of an understudied language family and
307
308
309
310
311
Ibidem,
Ibidem.
Ibidem,
Ibidem,
Ibidem,
p. 268.
p. 269.
p. 276.
p. 277.
248
MARK TURIN, BENJAMIN CHUNG
help to define the conceptual and socio-linguistic boundaries and internal relations of
Tibeto-Burman languages.
Across the Tibeto-Burman language family, the ‘green-blue’ discrepancy manifests
in multiple languages (such as Kulung, Thulung Rai, Sunwar, Yakkha, Tibetan, and
Yolmo) and articulates with more recent work on the perception of colour. Nominalisation,
especially using the pan-Tibetan nominaliser <-po>, is common throughout the family,
and is widely although not universally attested in Limbu, Kulung, Tibetan, Ladakhi and
Thangmi through various different suffixes. Additionally, there is heavy borrowing in
colour terminology in the Tibeto-Burman language family, and loans from Indo-Aryan
(Hindi and Nepali) and more dominant Tibeto-Burman languages are widely attested in
the data we have presented here.
Outliers in the Tibeto-Burman family – in terms of colour terminology at least –
offer intriguing insights into the diversity of colour encoding in not only lexicon and
grammar, but in how understandings of colour are connected to philosophy, religion,
and culture. Through such readings, we learn that colour cannot be simply understood
through morphological analysis or by lexical comparison. Ladakhi in particular proves
challenging as its colour range is rich, conceptual and rooted in specific cultural aspects
and forms, and because individual colours attested in Ladakhi do not appear to coalesce
into clear categories as theorized by scholars in the field.
Ultimately, this survey serves as a basis for further inquiry into the rich area of colour
terminology in Himalayan languages, and it is our hope that future researchers will use
this contribution as a consolidated, typological reference to colour in the Himalayan
region. Furthermore, we hope to have brought timely attention to the internal diversity and
variety that exists within these languages, as many of these speech forms are endangered
with dwindling speaker population. To that end, any work that profiles and makes visible
less commonly-known and often poorly documented languages can be interpreted as
form of language revitalisation, however modest.312 In conclusion, we hope to have
made a modest contribution to an established and ongoing theoretical discussion, to
have helped challenge the universal application of an exciting theory by introducing data
from an important and under-represented language family. Only through careful description,
rigorous documentation and comparative analysis can we take out understanding of human
cognition and perception to a more nuanced place.
312 M. Turin, Language Endangerment and Linguistic Rights in the Himalayas: A Case Study from Nepal,
“Mountain Research and Development” (25) 1, (2005), p. 4–9.